The Ancient Quarrel between

Philosophy and Poetry Revisited

This page intentionally left blank
 THE ANCIENT QUARREL BETWEEN
PHILOSOPHY AND POETRY REVISITED

Plato and the Greek Literary Tradition

Susan B. Levin

OXTORD
UNIVERSITY PRESS

2OO1
Oxford University Press
Oxford New York
Athens Auckland Bangkok Bogota Buenos Aires Calcutta
Cape Town Chennai Dar es Salaam Delhi Florence Hong Kong Istanbul
Karachi Kuala Lumpur Madrid Melbourne Mexico City Mumbai
Nairobi Paris Sao Paulo Shanghai Singapore Taipei Tokyo Toronto Warsaw
and associated companies in
Berlin Ibadan

Copyright © 2001 by Susan B. Levin

Published by Oxford University Press, Ine.
198 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press.
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

Levin, Susan B., 1961-
The ancient quarrel between philosophy and poetry revisited :
Plato and the Greek literary tradition / Susan B. Levin,
p. cm.
Includes bibliographical references and indexes.
ISBN 0-19-513606-3
i. Plato. 2. Greek literature —History and criticism —Theory, etc.
3. Language and languages —Philosophy.
4. Onomastics. 5. Poetics. I. Title.
B395-L375 2000
184 —dczi 99-054667

9 8 7 6 5 4 3 2 1
Printed in the United States of America
on acid-free paper
For Barry
This page intentionally left blank
 ACKNOWLEDGMENTS

I have had occasion to present ideas developed in the book in various settings. Some
material was presented at several of the annual conferences cosponsored by the
Society for Ancient Greek Philosophy and the Society for the Study of Islamic
Philosophy and Science, which are held each October at SUNY Binghamton.
Portions of it were also presented to the Department of Philosophy at SUNY
Albany, at the American Society for Aesthetics conference at Asilomar in 1997,
and at the Conference on Literature and Ethics, which took place at the Uni-
versity of Wales, Aberystwyth, in 1996. Having the opportunity to present my work
in these venues helped to enrich my thinking about topics treated in the book.
My early work on the book was fostered by a Jean Picker Fellowship from
Smith College for 1995-1996, for which I am grateful, and the continuation of
the project was promoted by a leave from the college the following year.
My deep thanks go to Julius Moravcsik for valuable feedback throughout the
process. The book has benefited greatly from his attention to it. I am grateful to
Justina Gregory for reading the manuscript in draft form and for making nu-
merous helpful suggestions. I wish to thank Alan Code, Gail Fine, John Malcolm,
Andrea Nightingale, and Charles Young for their insights regarding issues treated
in the book. I am also grateful to the referees for Oxford University Press for very
helpful comments. In addition, I appreciate the kindness of Glenn Most, who
put at my disposal a copy of Studies on the Derveni Papyrus, ed. A. Laks and
G. W. Most (Oxford, 1997) in advance of the book's general availability. Thanks
are due to Peter Ohlin and Robin Miura of Oxford University Press for their
assistance and expertise. Finally, I am indebted to my husband, Barry Lustig,
for his extraordinary support during the entire process. It is to him that the book
is dedicated.
I gratefully acknowledge the permission of Ancient Philosophy and Phoenix:
Journal of the Classical Association of Canada to include previously published
material. Also included, by permission, is material from Classical Philology,
Vlll ACKNOWLEDGMENTS

published by The University of Chicago Press. © 1997 by The University of Chi

cago. All rights reserved. Finally, I have drawn from an essay that appeared in
Feminism and Ancient Philosophy, ed. Julie K. Ward. © 1996 by Routledge.
Material included by permission of Routledge, Inc.

Northampton, Massachusetts S. B. L.
January 2000
 CONTENTS

ABBREVIATIONS AND TEXTS XI

INTRODUCTION 3

1. LITERARY ETYMOLOGY 13
Appendix: Other Expressions of a Belief in Substantial Ties
between Onomata and Elements of Reality 31

2. ETYMOLOGY AND THE CRATYLUS' SOURCES:

AN INVESTIGATION OF THE LITERARY TRADITION'S IMPORTANCE 42
1. Introduction 42
2. Orthotes Onomaton: The Cratylus' Engagement
with Literary Practice 45
3. Other Proposed Sources of the Cratylus' Discussion
of Etymology 64

3. THE STRUCTURE AND OUTCOME

OF THE CRATYLUS' INQUIRY 80
1. Introduction 80
2. The Techne Issue in the Cratylus 82
3. Nature and Convention: The Cratylus' Resolution 89
4. The Dialogue's Conclusion 94

4. CONCEPTIONS OF APPROPRIATENESS: PLATO'S

REVISION OF LITERARY USAGE IN THE PHAEDO AND REPUBLIC 5 99
i. Pouring New Wine into Old Skins: Reflections on Literary
and Platonic Eponymy 99
X CONTENTS

2. Functional Terms: Literary Usage and Plato's

Revision Thereof 119
3. Conclusion 126

J. THE QUARREL BETWEEN PHILOSOPHY AND POETRY:

A REEXAMINATION OF POETRY'S ROLE IN PLATO'S REPUBLIC 12J
1. Introduction 127
2. Poetry and the Concept of Techne: The Cratylus
Briefly Revisited 129
3. The Republic's Challenge to Poetry's Techne Status 131
4. Plato's Defense of Philosophy as the Techne
Par Excellence 135
5. A Techne of Poetry? The Status of Revised
Literary Practice 143
6. Poetry's Function in the Ideal Polls 150

CONCLUSION 168

BIBLIOGRAPHY 173

GENERAL INDEX 189

INDEX LOCORUM 204

 ABBREVIATIONS AND TEXTS

Abbreviations
Aeschylus (Aesch.) Hec. Hecuba
Ag. Agamemnon Hel. Helen
Cho. Libation Bearers Hipp. Hippolytus
Eum. Eumenides IA Iphigenia in Aulis
Pers. Persians IT Iphigenia in Tauris
PV Prometheus Bound Med. Medea
Sept. Seven against Thebes Or. Orestes
Supp. Suppliants Phoen. Phoenician Women
Rhes. Rhesus
Aristotle (Arist.) Supp. Suppliants
Gael. De Caelo Tro. Trojan Women
DA De Anima
EE Eudemian Ethics Herodotus (Hdt.)
EN Nicomachean Ethics
GA Generation of Animals Hesiod (Hes.)
HA History of Animals Op. Works and Days
Metaph. Metaphysics Th. Theogony
Rhet. Rhetoric
Top. Topics Homer (Horn.)
II. Iliad
Euripides (Eur.) Od. Odyssey
Ale. Alcestis
Andr. Andromache Pindar (Find.)
Bacch. Bacchae Isthm. Isthmian Odes
Cyc. Cyclops Nem. Nemean Odes
El. Electra Ol. Olympian Odes
HF Heracles Pyth. Pythian Odes

XI
Xll ABBREVIATIONS AND TEXTS

Plato Rep. Republic

Chrm. Charmides Soph. Sophist
Cra. Cratylus Symp. Symposium
Criti. Critias Tht. Theaetetus
Euphr. Euthyphro Tim. Timaeus
Euthd. Euthydemus
Grg. Gorgias Sophocles (Soph.)
La. Laches Aj. Ajax
Leg. Laws El. Electra
Phd. Phaedo OC Oedipus at Colonus
Phdr. Phaedrus OT Oedipus Tyrannus
Pol. Politicus Phil. Philoctetes
Prm. Parmenides Trach. Women ofTrachis
Prt. Protagoras

DK H. Diels. [1951-1952] 1992. Die Fragments der Vorsokratiker.

Ed. W. Kranz. Vols. 1-2. 6th ed. Reprint, Zurich: Weidmann.
LSJ H.G. Liddell and R. Scott. 1940. A Greek-English Lexicon, gth ed.
Revised and augmented by H. S. Jones. With a revised supplement,
1996. Oxford: Clarendon Press.
N2 A. Nauck, ed. 1889. Tragicorum Graecorum Fragmenta. 2d ed.
Leipzig: Teubner.
OCT Oxford Classical Text

Texts
Citations of Plato, Homer, Hesiod, Pindar, Aeschylus, Sophocles, Euripides, and
Herodotus are from the OCT editions. For volume i of Plato, I have used the
1995 edition of Duke, Hicken, Nicoll, Robinson, and Strachan. For authors other
than those mentioned above, I have used the texts listed in the Bibliography.
Unless otherwise noted, translations of Plato are my own, and renderings of
Homer, Hesiod, Pindar, the tragedians, and Herodotus are from the following
sources, at times with modifications (publication details are provided in the Bib-
liography): Iliad, R. Lattimore; Odyssey, R. Fitzgerald or A. T. Murray; Hesiod,
S. Lombardo or H. G. Evelyn-White; Pindar, F. J. Nisetich; tragedy, D. Grene
and R. Lattimore, eds., The Complete Greek Tragedies (except that renderings
of Sophocles' Oedipus at Colonus are from R. Fitzgerald, and those of Euripides'
Iphigenia in Aulis are either from The Complete Greek Tragedies or from P.
Vellacott); Herodotus, A. de Selincourt; and the Homeric hymns, A. N.
Athanassakis. Unless otherwise indicated in the Bibliography, translations of
French and German sources are my own.
 The Ancient Quarrel between
Philosophy and Poetry Revisited
This page intentionally left blank
 INTRODUCTION

Plato's middle dialogues have received an extraordinary amount of attention,

and much work has been done to illuminate his engagement with other phi-
losophers and the sophists as he articulates a range of doctrines therein.1 Socrates,
Parmenides, and Heraclitus are regularly identified as the key influences on
Plato's theory of Forms. According to the standard account, Plato continues
Socrates' quest—undertaken in opposition to the sophists —to articulate what
objective standards in the arena of values consist in, and adds to this emphasis
on values a strong concern with other entities, particularly those in the domain
of mathematics. In addition, as Aristotle emphasizes, Plato departs from his
teacher by insisting on the separate existence of the objective standards in ques-
tion.2 Moreover, he is in substantial accord with Parmenides' depiction of real-
ity as a changeless whole, although Plato's insistence on the necessary existence
of Forms in a range of cases evinces his refusal to endorse his predecessor's strict
construction of the principle that what is, is one. 3 Finally, Plato takes up

1. As Kahn (1996,44) points out, while the term "middle dialogues" originally referred to "the
stylistically intermediate group," it is currently applied to the works of Plato's "middle period"
delineated with regard to content (i.e., to those dialogues that introduce and articulate the theory
of Forms). As my remarks in what follows make clear, by "middle dialogues" I have in mind those
that are intermediate in the latter respect. The Phaedo, Symposium, and Republic fit closely to-
gether as dialogues that present various facets of the theory of Forms; on the Cratylus member-
ship in this group of dialogues, see n. 4.
2. See Phd. y8d, loob; Symp. 2iib; Rep. 5i6b; Prm. 1286-1293, i2gd, ijob-d. In Aristotle, see
EN io96b; Metaph. gSyb, io86b.
3. For Parmenides identified as "the most important single influence" on the theory of Forms,
see Kahn 1996, 82. Curd (1998, 228-41) too stresses the centrality of Parmenides as a backdrop for
Plato's account. She diverges from the dominant interpretation of Parmenides' monism, how-
ever, in claiming that his "arguments allow for a plurality of fundamental, predicationally uni-
fied entities that can be used to explain the world reported by the senses" (241).

3
4 THE ANCIENT QUARREL

Heraclitus' emphasis on the striking presence of change but restricts the appli-
cability of that account to the spatiotemporal world, insisting that reality, or nature
proper, is utterly free of mutation.
In the area of epistemology, Plato endorses Parmenides' contention that
knowledge, or understanding, has an exalted status, but his way of positioning
opinion solidly between it and ignorance (Rep. 476-480) makes clear that he is
contesting Parmenides' insistence that all cognitive states besides knowledge are
entirely without value. In addition, Plato's delineation of the role of sense-
perception in recollection (Phd. 72-77) constitutes a refinement of accounts,
offered by various Presocratic philosophers, that endorse perception's ultimate
subordination to thought without stating clearly the role of each and how the
two faculties interrelate. Moreover, the middle dialogues' description of what
knowledge involves, and how it is acquired, poses a challenge to the sophistic
view that the highest possible cognitive state is opinion.
Regarding ethics, the form and direction of Socrates' concern is a crucial
inspiration to Plato, while sophistic relativism is targeted quite forcefully. In
addition, certain Presocratics (above all, Democritus and Heraclitus) have a lim-
ited role as antecedents with respect to the general direction of Plato's theo-
rizing. Commentators have recognized, moreover, that the Greek literary tra-
dition, in particular Homer and tragedy, is a central target of Plato's treatment
of human character in a discussion (Rep. 2-3,10) that interweaves ethical and
pedagogical issues. Aside from this recognition of Plato's concern with litera-
ture in the arena of values, interpreters have taken his pivotal antecedents and
opponents as he develops key doctrines in the middle dialogues to be philo-
sophical and sophistic.
Plato's salient interests in the middle dialogues, however, extend beyond these
arenas to encompass topics in the domain of the philosophy of language. In fact,
an examination of the middle dialogues reveals a strong concern with linguistic
issues, in particular with the correctness of words or names (orthotes onomaton).
This interest is prominent in the Cratylus, where Plato concentrates on etymol-
ogy, rejecting the view that it is either a fruitful guide to essences or a linguistic
counterpart to the right conception thereof; in the Phaedo, where, seeking a lin-
guistic correlate to his account of reality centering on Forms, Plato turns to the
eponymy, or "named-after," relation; and in Republic 5, where his view of real-
ity leads Plato to occupy himself with the proper application of what one may
call "functional terms" (e.g., "ruler," "parent").4 In what follows, I investigate

4. With regard to chronology, I am in accord with those who place the Cratylus near the
beginning of that group of dialogues comprising the Symposium, Phaedo, and Republic: for this
view, see Kahn 1975,1986; Ross 1955; Meridier 1931; von Arnim 1912. Although some have linked
it with the Theaetetus, Luce (1964) and Calvert (rgyo) provide a range of grounds for adhering to
the earlier date. Cf. Brandwood's study of Platonic chronology (1990), which groups the Cratylus
with the Phaedo and Symposium, and Kahn 1996. With respect to thematic considerations, while
final determinations are of course not possible, in my view Plato's handling of orthotes onomaton
 INTRODUCTION 5

all three discussions with the aim of supplementing familiar accounts of the
key antecedents of Plato's reflections. I will argue, based on literary evidence
and consideration of the dialogues, that Plato's engagement with this tradition
in the middle period—when he articulates many of the theories for which he
is best known — plays a more important role in his development than has been
recognized.
Regarding the Cratylus' handling of orthotes onomaton qua etymologizing,
interpreters' accounts of sources have focused on philosophers and sophists. In
contrast to previous commentators, I maintain that Plato's most direct and promi-
nent opponent with respect to etymologizing is in fact the Greek literary tradi-
tion of the eighth through fifth centuries. 5 My claim is not that other proposed
sources (e.g., Heraclitus and Protagoras) are irrelevant to Plato's reflections in
the dialogue but rather that they are not justifiably pinpointed as the most cen-
tral targets of his critique. In order to ground this contention about literary prac-
tice, chapter i concentrates on authors' powerful interest, manifested in their
recourse to etymology, in what they view as deep connections between elements
of language and of reality. In their conduct of this activity, as we will see, they
focus most often on showing how a wide range of proper names, once analyzed,
disclose something salient about the natures of their individual bearers. This
extensive survey of literary etymology, including the delineation of categories
and subcategories into which authors' analyses fall, provides an indispensable
basis for the consideration of Plato's critique of etymology in the Cratylus.
Based on literary evidence and what transpires in the dialogue itself, I con-
tend in chapter 2 that the literary tradition is the etymological discussion's most
direct and central opponent. The Cratylus' references to Protagoras and Prodicus,
conjoined with their known interest in orthotes onomaton, have appeared to many
to provide solid evidence for the claim that the dialogue bears the marks of a
strong sophistic influence. Although this view of the sophists' impact may seem
plausible at first glance, it does not survive scrutiny. Once one sees how Prodicus
and Protagoras handled orthotes onomaton, and that Plato's interest takes a very
different form, it becomes evident why sophistic reflections have only limited

and metaphysical issues pertaining thereto, particularly as it involves the Cratylus and Phaedo,
offers strong support to the earlier placement. It should be noted, however, that my thesis about
Plato's relationship to the literary tradition does not depend on the outcome of the debate about
chronology in the sense that the scope and depth of his engagement with it remain the same re-
gardless of where one chooses to locate the Cratylus.
5. With respect to chronology in the case of Homer and Hesiod, I follow Kirk (1989, 6—7),
who proposes a date for Homer in the middle to late eighth century and places Hesiod in the
early seventh. The figures considered in the present study of etymology and eponymy are Homer,
Hesiod, Pindar, Aeschylus, Sophocles, Euripides, and Herodotus. When exploring the use of
functional terms, I concentrate on tragedy. "Literary tradition" is preferable to "poetry" as a way
of referring specifically to Plato's sources in the cases of etymology and eponymy because the
former rubric, but not the latter, allows for the inclusion of Herodotus.
6 THE ANCIENT QUARREL

bearing on the Cratylus. Having offered support to the view that Plato's recourse
to the literary tradition is strongly evident there, I supplement my account of its
centrality to his assessment of etymology by arguing that the Cratylus is not the
only place where he evinces a concern with the way in which literary authors
exploit the descriptive content ofonomata. When discussing early education in
Republic 2-3, Plato dwells extensively on the content of acceptable poetry. Most
important for our purposes is the fact that he addresses explicitly the topic of
names. Following an investigation of the Cratylus' ties to the literary tradition,
I propose that if one considers its etymologies of individual deities' names in
light of the Republic's criteria for acceptable poetry, a highly significant pattern
comes to light. In the final section of the chapter, I complete my discussion of
the Cratylus' sources by commenting on candidates who found favor as Plato's
inspiration in the nineteenth and twentieth centuries. Having already discussed
Prodicus and Protagoras due to Plato's mention of them when framing the
Cratylus' inquiry into orthotes onomaton, at this juncture I consider a range of
additional figures—above all, philosophers—suggesting why accounts that as-
sign them a central place are also open to question.
Chapter 3 focuses on the Cratylus philosophical concerns and outcome. In
the early and middle dialogues, Plato is consistently occupied with two closely
related questions: what entities can properly be said to be real (i.e., have natures),
and who may justifiably be called eudaimon (i.e., said to flourish)? Since one
must practice certain technai in order to determine what is unqualifiedly real,
and knowing what has this status is necessary for flourishing, one can understand
Plato's qualifications for being eudaimon only if one examines his views about
techne. Plato's systematic critique of poetry in dialogues up to and including
the Republic transpires in terms of challenges to its techne status. Early in the
Cratylus, Plato "grants" the rank of techne to the practice of naming qua ety-
mologizing. As in the Gorgias, however, where Plato investigates rhetoric, in this
case too what purports at first to be a techne later turns out not to meet his strict
criteria. In the Cratylus, the literary tradition provides Plato with crucial mate-
rial employed in his challenge to the techne status of naming; by the time the
dialogue draws to a close, however, Plato has rejected that tradition's methodol-
ogy and its guiding assumptions. While interpreters have noted Plato's inclusion
of a "craft analogy" early in the Cratylus, when the discussion turns briefly to
shuttles, they have not recognized the crucial role played by the concept of techne
in the dialogue as a whole. In this chapter I show how Plato frames and executes
his challenge to the view that naming qua etymologizing qualifies as a techne.
Interpreters diverge substantially in their views not only about the Cratylus'
sources but also regarding its content and significance. Some have doubted that
the dialogue has a serious purpose and concluded that it is of only minor philo-
sophical importance. Others, in contrast, attending to the opening portion of
the Cratylus (383—390) and to Plato's handling of etymology or phonetic consti-
tution, maintain that the dialogue advances a theory that one may treat as Plato's
own. While I agree that Plato's intentions in the Cratylus are serious, in chap-
 INTRODUCTION 7

ter 3 I argue that they do not issue in the elaboration there of substantive posi-
tive views. Instead, after a lengthy inquiry, Plato renounces constructions of
naturalness and appropriateness that have as their foundation the constitution
of ononwta. Although the dialogue has a largely critical focus, Plato offers clues
to the direction of his positive theorizing. Discussion of these leads naturally
to reflection on Plato's preferred approaches to orthotes onomaton. This is the
focus of chapter 4, which examines the Phaedo and Republic 5. While the topic
of the Cratylus' sources has received significant attention, the issue of ante-
cedents has not been a locus of concern among interpreters of the Phaedo and
Republic 5. With respect to the latter two cases, I argue that Plato's own theo-
ries of orthotes onomaton constitute revised versions of notions that figure
importantly in literature.
In the Phaedo, Plato uses the framework of eponymy to explain how terms
such as "just" and "beautiful" may properly apply both to Forms and to their
participants in the spatiotemporal realm. Although commentators have seen the
importance of eponymy to Plato's thought, they have not examined its anteced-
ents. Based on the study of extant evidence, I contend that the literary tradition's
handling of eponymy, which, like that of etymology, may be divided into
numerous categories and subcategories, constitutes a heretofore unexplored
precedent for the new, technical usage that correlates with Plato's view of real-
ity. Most importantly, reflection on literary eponymy allows one to distinguish
Plato's inheritance from those innovations that typify his employment of the
device with a philosophical end in view.
In the second portion of this chapter, I turn to Republic 5, where Plato's
theory of the assignment of sociopolitical and familial roles in the ideal com-
munity leads him to grant an important role to a revised version of a third lin-
guistic device central to Greek literature. Plato attends seriously to the pre-
scriptive use of terms like "mother" in tragedy and incorporates in the Republic
the notion, stressed repeatedly by the poets, that biological ties are neither a
sufficient nor even a necessary condition for the ascription of kinship bonds.
In addition, like the tragedians, he distinguishes explicitly between being
someone's relation in name only—that is, merely bearing the labels of kin-
ship—and behaving in the way characteristic of one with a given role. Far from
simply incorporating the tragedians' views wholesale, however, Plato's handling
of functional terms in Republic 5 reflects key points of divergence from liter-
ary practice. Having investigated Plato's revised treatments of orthotes onomaton
with respect to eponymy and the use of functional terms, I end chapter 4 with
remarks on a possible tie between his constructions of these devices insofar as
individuals correctly dubbed "rulers" are regarded as such based on their aware-
ness that key terms have both primary and secondary referents (i.e., Forms and
their participants, respectively).
From our perspective, according to which a wide range of disciplinary bound-
aries are assumed as a point of departure for inquiry—even if one proceeds to
go beyond them —it may seem odd, even quite implausible, to claim that Plato
8 THE ANCIENT QUARREL

makes substantial use of a literary backdrop as he develops his own theories. Once
one examines the situation from the vantage point that Plato would have had,
however, this state of affairs appears much less remarkable. From this perspec-
tive, in fact, it would seem odd if Plato had ignored this group of influential
authors even, indeed precisely, where their reflections were directly relevant to
his own discussions.
Regarding the context in which Plato operates, the key point to stress is that
he and his audience were familiar not only with the work of philosophers and
sophists but also, and deeply so, with the constructions of literary figures. Marrou
(1981,25) emphasizes the preeminent role played by Homer, particularly the Iliad,
in classical education. 6 Plato himself is quite clear on the matter of Homer's
importance.7 Over time, Hesiod, Pindar, and the tragedians Aeschylus, Sophocles,
and Euripides were also recognized as influential educators.8
Plato's writings give eloquent testimony to the tremendous influence of the
traditional literary education. As Marrou (1981,118) observes, they illustrate "in
a striking manner the extent to which Plato's personal cultivation had been fos-
tered by and had profited from the traditional poetic education: quotations from
Homer, the lyric poets, and the tragedians come spontaneously to his pen." In
addition, Plato's discussions of literature in the Republic and Laws leave no doubt
about the role this tradition played in Athenian education: the preface to his
criticisms of poetry in Republic 2 reports as common knowledge the fact that
literary study occupied a central position in the traditional curriculum (37662-
9); on its importance, the Athenian Stranger's remarks in the Laws (8ioe6-8iiay;
cf. 65436-7) are also unequivocal.9
It is not merely some Athenian elite who had significant exposure to litera-
ture. The greatest of the dramatic festivals, the annual City (or Great) Dionysia

6. As Kirk (1989, i) observes, "learning his poetry by heart was an essential part of ordinary
education." On the topic of memorization, see the comments of Niceratus in Xenophon's Sym-
posium (3.5).
7. See, e.g., Rep. 60661-60733. The poet's significance is also evident from the force of the
Republic's attacks on him, as well as from the great number of passages in the dialogues in which
Plato either refers directly to Homer or clearly has him in view.
8. It is important to note that poets' writings had an educational impact in multiple contexts-,
on the one hand, the works of poets like Homer were introduced in primary schools; on the other
hand and more generally, the influence of these authors and their writings pervaded Athenian
culture, notably through the presentation of tragedies at the dramatic festivals. Regarding the
genesis of the schools, see Marrou 1981, pt. i, chap. 4.
9. On this issue, see also Prt. 324dy-326b6. In his attempts qua philosopher to propound what
he takes to be the optimal curriculum, Plato is rivaled by Isocrates, who insists that his activity
constitutes philosophy and that he is the genuine philosopher (Antidosis 270-71; cf. 285). Vari-
ous comments of Isocrates, taken together, both acknowledge the pervasiveness of poetry and
recommend its continued inclusion in general education. Like Plato, however, Isocrates advo-
cates a course of selective exposure and study and offers his share of criticisms (for his attitude,
see Antidosis 266-68; To Demonicus 51-52; To Nicocles 2-3, 13, 43-44, 48-50; Busiris 38-40; cf.
Panathanaicus 17-34).
 INTRODUCTION 0

held in Athens, became important in the sixth century.10 In the fifth and fourth
centuries, this festival, along with the Country Dionysia and Lenaia, saw the
performance of vast numbers of tragedies and comedies. In the fifth century,
original tragedies presented in Athens could be reproduced subsequently at
the Country Dionysia. During the fourth, repeat performances were allowed
even at the City Dionysia, and eventually it was stipulated that a certain pro-
portion of the festival would be occupied by the reproduction of classics from
the writings of Aeschylus, Sophocles, and Euripides. Large portions of the
Athenian population attended these dramatic productions, and full citizens
on the deme register received a subsidy (theorikon) from deme officials; this
apparently covered not just the cost of the ticket but also helped citizens to
meet other festival expenses.11
On the level of specific poets, Aeschylus enjoyed such popular acclaim that
following his death the Athenian people passed a decree allowing the produc-
tion of his plays to continue. Moreover, after the disastrous Sicilian expedition
during the Peloponnesian War, some Athenian prisoners received food and water,
and in certain instances their freedom, by reciting from memory excerpts from
the plays of Euripides. 12 On the comedic front, Aristophanes' Frogs was so well
received that it was performed a second time due to popular demand. 13
In fact, the Frogs itself illustrates the degree of acquaintance with poetry that
could be assumed even in the case of those ordinary Athenians who filled the

10. The traditional date of the first tragic performance is 534 B.C. under Thespis. My discus-
sion of the dramatic festivals is based on Pickard-Cambridge 1988, chaps, i, 2, 6; Parke 1977, pt. i,
chaps. 6, 7, 9. Pickard-Cambridge (1988, 58) ties the City Dionysia's achievement of importance
to the policy of Pisistratus. He claims, in addition, that the Didaskaliai included in Diogenes
Laertius' catalogue of Aristotle's writings "no doubt took its title from the official language of the
festival. The poet was said 8i8ac5Keiv ipayraStav or KG>ufi>8tav" (1988, 71); this terminology high-
lights poets' educational role.
11. During the City Dionysia, a holiday was declared (prisoners were even released on bail
for the occasion). Plato has harsh words for those "lovers of sights and sounds" who rush to attend
all of the dramatic festivals (Rep. 47$d); on the public's familiarity with dramatic performances,
cf. Isocrates, Panathenaicus 168. The privilege of a seat of honor (proedria) was granted by the
state, and certain portions of the theater were reserved (e.g., for the Council). Due to the festival's
tremendous popularity—more Athenians and foreigners gathered in the city for this event than
at any other time during the year—Athenian officials chose the festival as the setting for various
important displays. In the fifth century, the City Dionysia was set as the date for delivery of the
empire's tribute to Athens, and this tribute was actually displayed to the festival audience before
the tragedies were performed; in addition, there was an event centering on orphans of those who
had died in battle. In the fourth century, awards of golden crowns to politicians or other Athe-
nian benefactors were announced at the new tragedies. Moreover, the public importance of the
City Dionysia is suggested by the fact that directly following the event the Assembly held a meet-
ing in the Dionysiac theater at which officials' conduct of the festival was evaluated. On citizens'
role as chorus members, see Pickard-Cambridge rg88, 75-77.
12. In addition to Pickard-Cambridge 1988, 276, see Havelock 1986, 94.
13. The play was produced at the Lenaia in 405 B.C., where it took first prize. On the issue of
its repetition, see Pickard-Cambridge 1988, 275.
10 THE ANCIENT QUARREL

comedic audience.14 Aristophanes is quite aware of his own educational role and
of tragedy's importance to the public.15 In this play, he stages an Underworld con-
test between Aeschylus and Euripides, the goal being for each of the two poets to
show why he is of the greatest potential benefit to his native Athens and hence
should be permitted to depart from Hades in order to aid that city in its darkest
hour. In his conduct of the dispute, which ends with preference being granted to
Aeschylus as the superior educator, Aristophanes presupposes a more-than-generic
familiarity with the compositions of both tragedians.16 Regarding the Frogs' han-
dling of tragedy, Jaeger (1945, 377) observes that the play, "written in the brief in-
terval between the death of the two poets [Euripides and Sophocles] and the fall
of Athens, is charged with a ... tragic emotion. As the situation of Athens became
more hopeless, and the pressure on the morale of her citizens increased to the
breaking-point, they grew more eager for spiritual comfort and strength. Now at
last we can see what tragedy meant to the Athenian people."
Based on such well-known facts about the content of Athenian education-
viewed as occurring both outside and within the primary schools—one may as-
sume that while they themselves are neither poets nor part of the poetic treatment
of the issues in question, both Plato and his intended philosophical audience were
quite familiar with this literature and the evolving, often mutually critical dialogue
contained therein. This means that while Plato frequently makes explicit refer-
ence to poets, individually or in various collections, it is not necessary for him to
do so when alluding to or drawing on them, just as it is not required that he men-
tion other philosophers by name in order for his audience to be aware that he has
their ideas in view. Our different perspective on disciplinary boundaries, combined
with the fact that Plato's recourse to these sources is not always explicit, helps to
explain why commentators so often overlook their full importance. Since they
assume Plato's frequent engagement with philosophers and sophists, they regu-
larly seek and find evidence of that concern even where it remains unstated.
Democritus, for example, is never mentioned by name in the dialogues, but this
has not prevented his being identified as an important Platonic opponent.17 Be-

14. The likely date of the first comic performance is 486 B.C. (see Pickard-Cambridge 1988,
72; Parke 1977, 129). As Dover (1970, xiv) observes, "the paradox of comedy is that it was written
for a mass audience by sophisticated poets with high technical standards."
15. On the atmosphere in which the Frogs was produced and the play's import, see Jaeger
1945, bk. 2, chap. 5.
16. Both the Chorus's reaction to the decision (1482-84) and Pluto's final instructions to the
poet (1500-1503) stress Aeschylus' educational function.
17. See, e.g., Guthrie 1975, 37. To give another, less dramatic, example, although the Repub-
lic does not mention Parmenides, he has been treated as a central antecedent of the figures of the
Sun, Line, and Cave in Books 6-7: according to Ferguson (1963, 191), the passage "is highly
Parmenidean, and this fact is necessary to the understanding of it." In addition, as is frequently
observed, Republic 5'$ treatment of knowledge and opinion (474-480) has a strong Parmenidean
backdrop. On this Eleatic philosopher as a foundation for Diotima's "vision" in the Symposium,
see Kahn 1996, 69. My remarks at the outset of the introduction include further examples.
 INTRODUCTION 11

cause interpreters do not expect a similar level of involvement with literary sources,
however, they are less likely either to give explicit references their due or to dis-
cover evidence of a concern with them where such references are lacking. Study
of the relevant literary material helps to place one in the position in which Plato's
own audience found itself when confronted with his treatments of orthotes
onomaton; that is to say, it provides one with crucial background which that philo-
sophical audience would have appealed to without hesitation but which we, nearly
two and a half millennia removed from that setting, generally do not consult.
There is a noteworthy paradox at work here. Firm boundaries between phi-
losophy and science or philosophy and literature are not in place in Plato's time.
Plato is concerned, particularly in the Republic, to defend some crucial distinc-
tions between the latter two domains. Even as he works hard to forge what we
would call "disciplinary" boundaries, however, Plato remains very much a part
of his culture in the sense that, as I will argue regarding his philosophy of lan-
guage, in formulating his theories he continues to find it just as natural to draw
on literary sources as on philosophical and sophistic ones.
I should say a few words at this juncture about my own interpretive stance
toward Plato. Commentators are paying increasing attention to the "literary" or
"poetic" side of his thought and expression. Quite often this involves stressing
what they view as the central role in Plato's writings of literary devices such as
myths and allegories and using the fact that he wrote in dialogue form as a basis
for challenging what commentators have labeled a "doctrinal" stance toward the
interpretation of his writings.18 The alternative methodology endorsed might be
aporetic, or it might consist in the attempt, made frequently in recent work on
Plato, to forge middle ground —a so-called third way—between doctrinal and
aporetic stances.19 The present study shares with the work of many recent inter-
preters a concern with Plato and the "poetic" or "literary." It underscores quite
different manifestations of his interest, however, offering extensive support to
the view that one may illustrate the scope and depth of Plato's engagement with
literature while continuing to espouse a doctrinal approach, indeed, that attend-
ing to literary sources makes a crucial contribution to our understanding of his
theories. As is the case when one focuses on Plato's concerns in other areas, so
too with the relevant linguistic issues; one understands best what is distinctive
in his theories if one sets them against the backdrop of the pertinent anteced-
ents. As it happens, the most relevant backdrop with respect to orthotes onomaton
is the literary tradition, hence my concern with that tradition in the context of
pursuing a doctrinal methodology.

18. For recent accounts that emphasize such "literary" considerations, see Gordon 1999;
Rutherford 1995; Gonzalez 1995; Sayre 1995; Press 1993; Arieti 1991; Griswold 1988.
19. This phrase appears in the title of the collection of essays edited by Gonzalez that was
mentioned in the previous note. For his own latest attempts to articulate what adopting this ap-
proach involves, see Gonzalez 19983a, 1998b
12 THE ANCIENT QUARREL

Having investigated Plato's handling of linguistic issues in the Cratylus, Phaedo,

and Republic 5,1 end the book with reflections on his famous critique of poetry in
Republic 2-3 and 10. The Cratylus critical strategy with respect to etymology, whic
centers on the concept of techne, suggests a need and provides a blueprint for a
more comprehensive evaluation of poetry. As I will argue, while the former dia-
logue limits itself to one aspect of poetry's content (i.e., etymological analyses of
onomata that are supposed to disclose their referents' natures), the latter challenges
the techne status of poetry as such. Following an investigation of the ties between
these two critiques and of philosophy's status as the preeminent techne, I conclude
my treatment of the Republic with remarks on poetry's potential and on the rela-
tion envisioned by Plato between it and philosophy. Although Plato's insistence
that poetry may have a role in his polis only if its benefit can be demonstrated might
lead one to expect a strongly negative outcome to the Republic's inquiry with respect
to poetic praxis as a whole, I argue that his conclusions are in crucial respects
positive. My interpretation of Republic 2-3 and 10 diverges both from the view of
interpreters who claim that Plato excludes poetry from the community he envi-
sions and from the position of those who contend that it remains but falls strictly
in the domain of the "poet-philosopher." I claim instead that Plato reserves an
important role for a distinct practice of poetry both in the project of attitude for-
mation that is the focus of early education and in the broader communal context,
where poetic compositions will be integral to a range of civic occasions. Thus,
Plato's investigations otorthotes onomaton in the middle dialogues, conjoined with
the Republic's familiar critique of poets, offer strong support to the view that, in
his reflections on language and institutional structures, as in his theorizing about
reality and knowledge, Plato takes serious account of the most promising ideas
and talents that his predecessors—regardless of their identity—have to offer. Plato
is often seen as the source in Western thought of a dichotomy between philoso-
phy and literature that is highly pernicious. As I suggest in the book's conclusion,
if the interpretation defended here is on the right track, then this widespread view
of his role is in need of revision.
 1
LITERARY ETYMOLOGY

Writers in the literary tradition frequently articulate the descriptive content of

onomata, largely proper names, by way of etymology.' Thus, for example, they
trace "Apollo" to apollumi ("destroy") and "Hector" to echo ("protect"). Such
analyses of onomata are thought to disclose something important about their
referents' natures. Literary authors' reliance on etymology raises central issues
involving the link between elements of language and of reality. In what follows,
I explore literary practice in order to provide a basis for the examination of Plato's
critique of etymologizing in the Cratylus.
Previous accounts of the dialogue's sources typically focus on various philoso-
phers and sophists. One difficulty with such proposals is that, due to the general
paucity of surviving evidence regarding the views of earlier philosophers and
sophists, it remains unclear whether and to what extent various figures appealed
to etymology in the manner at issue in the Cratylus. In addition, even where a
certain amount of evidence exists, a failure to distinguish between multiple con-
structions otorthotes onomaton, or various types of "wordplay," has contributed
to a tendency to assimilate others' activity more closely to Plato's practice than
the existing relationship between their approaches justifies. The following survey
is therefore essential because it makes clear that the literary tradition's recourse
to etymology fits the bill as the most direct and central backdrop for the Cratylus
with respect to the extent and character of appeals to etymology, as well as con-
cerning the view of natures that grounds those appeals.

i. As Robinson (1969,101) has observed, in the Greek notion of an onoma "there lay undistin-
guished at least five notions that are distinct now: the proper name, the name, the word, the noun,
and the subject of predication." While literary etymology is not limited to proper names, these
are its most central concern. In addition, as we will see, these onomata art most directly perti-
nent as the foundation for Plato's inquiry in the Cratylus.

13
14 THE ANCIENT QUARREL

When one investigates literary etymology, what is particularly striking is the

wide range of criteria on the basis of which onomata are said to be assigned. First,
names might be traced to something important involving the birth and early
existence of their referents. Second, in numerous cases appellations are corre-
lated with features or aspects of their referents, among them character traits and
capacities. A third category comprises instances in which assignments are tied
to actions. Fourth, onomata might be linked to significant effects that their refer-
ents have on mortals. Fifth, writers offer etymologies of names based on objects
with which their bearers are associated. Sixth, a feature of the place from which
someone hails may ground an appellation. In a seventh type of case, a natural,
inanimate entity might receive its onoma based on an important individual with
whom it is connected. An eighth category consists of instances in which writers'
analyses of names underscore their referents' function in social or familial con-
texts. Ninth and finally, onomata may be traced to roles and special powers of
the divine. The foregoing list conveys a sense of the diversity of factors to which
literary authors appeal in drawing attention to the descriptive content of onomata.
In what follows, I provide a detailed exploration of writers' use of these criteria.2
First, as noted, appellations might be derived from something important
pertaining to an individual's birth and early life: more specifically, they may be
traced to the manner or direct cause of an individual's genesis, an object with
which the named individual is associated at birth, the place of an individual's
birth, events prior to birth, or those occurring not long thereafter. With respect
to the origin of individuals, Hesiod speaks of the deity whom gods and human
beings call "Aphrodite," "since it was in foam she was nourished" (OWEK' ev
difipra 9pe([>0r|, Th. 197-98). In Aeschylus' Suppliants, the king of Argos identi-
fies himself as Pelasgus, "son of Palaechthon earth-born" (toti yrvyevo'uq . . .
TlaXa.i'xQovoc, Iviq, 250-51); by so doing, Pelasgus highlights the descriptive
content of his father's name (cf. 255, %06vci). Another pertinent case involves
justice, which the literary tradition treats as a female deity: "In the fighting Orestes'
hand was steered by the very daughter of Zeus: Right we call her, mortals who
speak of her and name her well" (eGiye 8' ev |id%cu %ep6<; etr\m\ioc, Aioq Kopa,

2. No one set of categories could do justice to the complexity of literary etymology; the divi-
sions used here simply facilitate the organization of this material for present purposes. In the case
of passages that might reasonably be categorized in multiple ways, I make what seems on balance
to be the best choice. In addition, this survey is not intended to be exhaustive but simply to con-
vey a sense of the prominence and range of literary authors' interest in etymologizing. I should
note, moreover, that my interest here is not in the philological accuracy of writers' derivations
but solely in the analyses that they proffer and the criteria on the basis of which connections
between onomata are forged. Finally, and more generally, the book's investigation of literary treat-
ments of orthotes onomaton includes the study, facilitated by the Thesaurus Linguae Graecae
(copyrighted by TLG and the Regents of the University of California), of passages in which onoma
and related terms appear. With respect to literature, however, as in the case of Plato, my interest
is in the relevant linguistic phenomena, by whatever terms they are introduced.
 LITERARY ETYMOLOGY 15

, Aesch. Cho. 948-51).

Aeschylus traces the genesis of the name "Dike" to the fact that its referent is
Zeus's daughter (Atoq Kopec). Aeschylus' reference to her elsewhere as "Justice,
Zeus's virgin daughter" (fi Aioq nalc, TiapGevoq AIKT), Sept. 662) underscores this
same idea. Hesiod before him had spoken of the maiden as "virgin Justice, the
daughter of Zeus" (fi napQevoc, Alien, Aioq eicyeyama, Op. 256); given the simi-
larities between the formulations in Seven against Thebes and Works and Days,
it is conceivable, even likely, that Aeschylus intentionally recalls Hesiod.3 In
another pertinent case, Aeschylus asserts that ID'S son, Epaphus, received his
name based on the manner of his engendering by Zeus (PV 850-51): specifi-
cally, the poet says that Epaphus came to be as a result of Zeus's touch (etyayic,)
and that this mode of origin is reflected in his onoma (Supp. 40-48). Aeschylus
reiterates this etymology later in the play, noting that "Zeus by touch begot a
son" (Ze\)<; y' e^dmcop %eipi (|nTU£i yovov, 313) and that Epaphus is "truly named
from the laying on of hands" ("E7ia(|>o<;, dA,r|0a><; puoirov ejt(6vu|a.o<;, 3i5).4 In
Sophocles' Oedipus at Co/onus, in turn, the name of Parthenopaeus, Atalanta's
child, is analyzed as "Son-of-a-maiden" (E7ic6vu|j,O(; xr\c, 7ipoo0EV d8|a.f|TT|<; . . .
yovoq, 1321-22).5
In a different type of case, objects with which individuals are linked at birth
are the source ofonomata, as when Hesiod reports that Poseidon lay with Medusa
and that when Perseus cut off her head, a son emerged and was named "Chrysaor"
on account of the golden sword (ctop xpuoeiov) that he held as he sprang forth
(Th. 280-83). In other instances, the place of an individual's birth grounds
assignments. For example, Hesiod tells how gods and men call Aphrodite
"Cyprogenes because she was born on the surf-line of Cyprus" (Kimpoyevea 5',
on yevto TtepiKA.'UOTCp svi K.tmpcp, Th. 199).6 The poet also indicates that when
Perseus cut off Medusa's head, the horse Pegasus leapt forth; he received this
name due to his birth near the springs of Ocean ('QKeavov Ttapd TtTvyaq) (Th.
280-83).
For events prior to birth as the source of assignments, one may consult Pindar,
who recounts Heracles' prayer to Zeus that Telamon be granted a son, and his
instruction to Telamon to name his son "Aias" for the eagle (aietos) sent by Zeus
in answer to that prayer (Isthm. 6.49-53). Elsewhere, Herodotus asserts that the
Spartan king Ariston's son was called "Demaratus" due to his birth following

3. Both writers treat Dike as one whose aid mortals might hope for and enlist. Her role as
avenger is often stressed, as when the Chorus in Aeschylus' Libation Bearers says that Orestes was
assisted by Zeus's daughter in gaining retribution for his father's death (948-51). Concerning Dike's
function as avenger, see also Hes. Op. 213-73; Aesch. Sept. 662-73; Soph. A/. 1389-92.
4. Tr. Smyth (1922), slightly modified. See also Supp. 535, where Zeus is addressed a

5. Euripides mentions that Parthenopaeus is Atalanta's son (' £ii(&awr\q yovo<;) and links the
mother explicitly with Artemis (Phoen. 150-52); regarding the tie to Atalanta, see also Phoen. 1104-9.
6. See also Find. Pyth. 4.216, where the poet calls Aphrodite Kimpoyeveia.
16 THE ANCIENT QUARREL

the prayer (ard) of the entire Spartan people (demos) that Ariston have a son
(6.63.3). In both of these cases, others' hopes are the ultimate source of the ap-
pellations given: in the former instance, the eagle, represented in the name,
functions as the symbol of a wish fulfilled, while in the latter, the name's seman-
tic constitution reflects others' fervent hopes that its bearer be born.7
An onoma is traced to what transpires after birth in Pindar's remark that the
infant child of Evadne and Apollo "lay hidden . . . , his delicate body bathed in
the yellow and deep blue rays of violets (ux), from which his mother then named
him lamus, a name immortal forever" (O/. 6.54—57).8 Earlier in the ode, Pindar
links the onoma to a different source when he notes that, following Evadne's
abandonment of him, a pair of serpents fed the infant lamus blameless venom
(eOpeyavto ... d|j,e|j,(|>e! ia>) (O/. 6.45-47).9 In another, quite famous, example,
Oedipus is said to have received his name based on a misfortune that he suf-
fered soon after birth:

OEDIPUS: What ailed me when you took me in your arms?

MESSENGER: In that your ankles (jco6ff)v dp6pa) should be witnesses. . . . I loosed
you; the tendons of your feet (icoSoiv aK|xc(<;) were pierced and fettered. . . . From
this you got your present name. (Soph. OT 1031-36)

Elsewhere in the play Jocasta recalls that within three days of Oedipus' birth, "King
Laius pierced his ankles and by the hands of others cast him forth upon a pathless
hillside"
opoq, 718-19; see also 1349-50). In Phoenician Women, Euripides follows this treat-
ment of the name (25-27, 801-5; cf. 41-44). Finally, Herodotus claims that when
Eetion's son grew up, he was named "Cypselus" on account of that object, namely,
a chest (kupsele), which allowed him to survive an early danger (5.9281).
A second category of assignments is correlated with features or aspects of
individuals or groups. Here onomata may be tied to a physical feature or related
aspect of an individual (or common to a group of individuals), individuals' emo-
tional states, bearers' attitudes and character traits, or their skills and capacities.
Regarding the first subcategory, Homer asserts that Scylla was named based
on the sort of voice she has, namely, that of a newborn whelp (cjxovfi... OKtiXaKOi;
vEoyiAfjc;) (Od. 12.85-87). Hesiod claims that Earth's progeny Brontes, Steropes,
and Arges were nicknamed "Cyclopes" because they each had only one eye,

7. Since these assignments were made based on prebirth events and activities, they cannot of
course involve the bearer directly. In neither case, however, do considerations involving the actual
namer underlie the choice of onomata: in the former instance, Aias is named for Zeus's eagle,
but Zeus himself does not make the assignment; in the latter case, the Spartan people as a group
does not serve as namer, but rather the child's parents. I therefore group these passages with those
in which the bearer's origin and early life are at issue.
8. Pindar also calls Evadne a "violet-braided girl" (itavSa ioiAoKov, Ol. 6.30).
9. Nisetich (1980, 103) terms this "a kenning for honey."
 LITERARY ETYMOLOGY 17

which was set in the middle of their foreheads

Th. 144-45).
In another pertinent case, Aeschylus' analysis of the name "Parthenopaeus" is
based ultimately on the physical appearance of its bearer. In this connection,
the messenger in Seven against Thebes tells of the warrior's oath that he will
destroy Thebes and then of his appearance and demeanor (532-37); here the
poet stresses that Parthenopaeus' youthful and attractive exterior is at odds with
his ferocious attitude ("His spirit unlike his maiden name is savage," 6 8' c6(iov,
o-u TI 7iap0£vcov £7icQVU|K>v (^povrpct).10
Elsewhere individuals' emotional states ground assignments, as when Sophocle
has Aias agonize as follows regarding the connection between his name and his
current state of despair: "Agony. Who would have thought my name and for-
tune could square so well together! My name is Aias: agony is its meaning" (alar
tig dv TCOT' (B£0' o)S' Ejrobvuja.ov TOTJUOV S.'uvoiaeiv ovoua mlc, entile, Kaicoiq; vuv
yap Ttdpeoii KOI die, aid^Eiv E^oi, A;. 430-32)." In the Bacchae, Euripides
alludes to the descriptive content of the name "Pentheus," that is, its derivation
from penthos ("grief): in the midst of a hostile exchange with Dionysus, Pentheus,
having identified himself by name, is told by the god that he "shall repent that
appellation" (508).
A third and quite popular subcategory is that involving bearers' attitudes and
character traits. One may begin by consulting the Iliad, where Homer draws
attention to the semantic constitution of the name "Thersites" (from tharsos
[Aeolic thersos], here "rashness") by depicting its bearer's reckless attitude and
conduct (2.212-64).12 In the Odyssey, the poet observes that "Aias was truly lost
among his long-oared ships. . . . He would have escaped his doom . . . had he
not uttered a boastful word in great blindness of heart (|U£y' ddo0r|)" (4.499-
503).1? Later in the poem, Homer singles out for mention the suitor named
"Ctesippus," who is said to trust in his extraordinary wealth in the pursuit of
Penelope
20.288-90).
The tragedians offer relevant etymologies of various individuals' names. For
example, trying to persuade Philoctetes to join forces with the Greeks, Neoptolemus
tells him that "it is a glorious heightening of gain (KaX,r\ yap r| 'TUKTTICK;)," first

10. Although he does not refer directly to the bearer's onoma, Euripides too focuses on
Parthenopaeus' appearance by having Adrastus depict the warrior as "the son of huntress Atalanta,
Parthenopaeus, supreme in beauty" (6 Tfj<; KwayoiJ 8' aXkoc, 'AiaXavrrn; yovoq Ttaii; flapBEVonatoi;,
eiSoc; E^oxaraxtoi;, Supp. 888-89). Regarding the present subcategory, cf. Hdt. 1.139.
11. Later in the play, Teucer says the following to the dead Aias about his own emotional state:
"Oh, what a crop of anguish (OCTOK; (iviai;) you have sown for me in death!" (1005).
12. For discussion of the figure and onoma, see Nagy 1979, 259-62. See Kirk (1985,138-39) on
the significance of Thersites' standing in the Iliad as the only character for whom Homer omits
both patronymic and city or area of origin.
13. This connection is repeated in line 509.
18 THE ANCIENT QUARREL

to be healed and then to achieve great fame for the conquest of Troy (Soph.
Phil. 1344-47). Subsequently, Heracles gives a more specific account of what
Philoctetes stands to gain, placing greater emphasis on the prospective material
benefits (1413-33). While it is true that a desire for such advantages was widely
shared, the former passage, in which the descriptive content of Philoctetes' own
name is underscored, indicates that Sophocles envisions a specific relation be-
tween this individual and the trait in question. Interestingly, Euripides offers an
etymology of the name "Aphrodite" that differs significantly from Hesiod's when
he has Hecuba say that "all acts of human intemperance are Aphrodite, and the
goddess is rightly called by this name since it begins with the word for folly" (T
uxopa yap navi eoiiv 'A<j>po5iTri (3pOTOii;, Kai TOWO^' 6p0<5<; d<]>poo"6vr|<; dp%ei
0ea<;, Tro. 989-90).14 In the Bacchae, in turn, Pentheus' mother, Agave, is asked
by the Chorus whether she is proud (dydAAri;, 1197). Elsewhere, Euripides high-
lights the descriptive content of the name "Dolon" by emphasizing the bearer's
dolos ("cunning," "treachery," Rhes. 215; cf. 216-18). In this play, Euripides goes
to great lengths to forge parallels between Dolon and Odysseus—a hero renowned
for his cunning; one way in which he accomplishes this is by also associating
Odysseus with dolos, using an adjective formed from the noun (for the phrase
dolios Odusseus, see Rhes. 894). With respect to the present subcategory, one
may consult, in addition, Euripides' Iphigenia in Aulis, where Agamemnon
says, with himself in mind, "I am the son of Atreus. Do you think he shrinks
from your eye, Menelaus?"q
, 321).15
Finally, individuals' skills and capacities ground assignments, as is evidenced
by Odysseus' address to a man he has killed in battle ("Socus, son of wise
Hippasus the breaker of horses"; ro £(*>%', 'iTuidoo'u me Sa'i,'(|>povo<; irato5d|a,oi.o,
Horn. II. 11.450). Regarding another name with hippos as a component, a
messenger recounts the sudden panicking of Hippolytus' horses and their rider's
skill in the handling of them (Eur. Hipp. I2i8-2o);16 along these same lines,
Adrastus says that Hippomedon had one goal, namely, "by skill in hunting,
archery, and horsemanship (iTtTioicj), to train himself for useful service to his
city" (Eur. Supp. 885-87).17
Proteus and Psamathe originally named their daughter "Ido" based on her
physical resemblance to her mother, but when she was older, "they changed her
name to Theonoe, for she understands all things that are, all things to be, that

14. My translation; I follow Barlow (1986) in rendering td (jxopo here as "acts of human
intemperance."
15. Operative here, as in the case of Hector and his son, to be discussed below, is the assump-
tion that a son's nature, at any rate when viewed as commendable, will reflect that of his father.
Regarding this subcategory, see also Eur. IA 1402 (with 1410-11, 1422-25, 1595; cf. 1375-76) and
Hdt. 7.231.
16. Regarding the linkage of Hippolytus with horses, see also 307-10, 582-83, 1131-34.
17. Tr. Vellacott (1972).
 LITERARY ETYMOLOGY 19

divination alone can tell"

(Eur. Hel. 8-14).18 In a relevant instance from
the Odyssey, Alcinous sends for "our minstrel, Demodocus, whom the god made
lord of song, heart-easing"
, 8.43-45); m's analysis ties the name most directly to a special
capacity divinely bestowed. Elsewhere, Euripides offers an etymology of the name
"Thoas" based on the bearer's ability to move with tremendous speed (65 anew
7i68a TiGeii; voov niepolq ec,TOWOJO.'fjA,6e t68e TtoScoKeiaq %apiv, IT 32-33).
Both Hesiod and Aeschylus comment on the name "Prometheus." In the
Theogony, Hesiod says that Clymene bore "clever Prometheus, full of various
wiles" (ripouTjGea nomKov aioXojj.ri'nv, 510-11), and offers a prime example of
Prometheus' cunning in the tale of his attempt to deceive Zeus when Prometheus
divided a certain ox into portions (Th. 535-60); even though this incident put
Zeus on his guard, Prometheus subsequently "outwitted him and stole the far-
seen gleam of unwearying fire in a hollow fennel stalk" (565-67). In Works and
Days, Zeus addresses Prometheus as "son of lapetus, surpassing all in cunning
(TOVTCOV rcepi |j,f|8ea eiScac;)" (54). Additionally, in Aeschylus' Prometheus Bound,
Cratus taunts Prometheus about the latter's name: "The gods named you wrongly
when they called you Forethought; you yourself need Forethought to extricate
yourself from this contrivance" (\|/eu8<Bv6|axflc; oe 8aiuove<; IIponr|6ea KaKovaiv
amov yap oe 8ei npo^Qiaq, OTCO iporccp if|o8' eKK^XioGriorj xe%vr|<;, 85-87).
Hesiod tells, moreover, how Epimetheus failed to recall the advice of his
brother Prometheus never to accept a gift from Zeus but to return it lest it bring
evil to mortals. Instead, he accepted the gift and only gained understanding after
the evil had already transpired (Op. 83-89). Although Hesiod does not highlight
explicitly the descriptive content of the name "Epimetheus" ("Afterthought") and
the appropriateness of this onoma to its bearer, his account of Epimetheus' de-
ficiency performs an equivalent function. Pindar too draws attention to the
name's descriptive content, by using a synonymous adjective to characterize the
bearer, namely, opsinoos ("late of thought," Pyth. 5.28; cf. Hes. Th. 511, where
the term hamartinoos appears). In the same line, Pindar names Prophasis as
Epimetheus' daughter; the semantic constitution of her name (which means
"Excuse") is clearly based on that of her father's onoma.
Finally, upon his arrival on Ithaca, Odysseus asks his divine protector,
Athena —who is then disguised as a mortal—whether he has in fact reached his
island home. To this query, the goddess responds: "Always the same detachment!
That is why I cannot fail you, in your evil fortune, coolheaded, quick, well-spoken
as you are!" (cdei TOI TOIOTJTOV evi otrjGeooi vorpa' TG> oe Kai OTJ 8uvaum
TipoXureiv 8-uotr|vov eovTa, ouveK' ejtr|Tf|<; eooi Kai dy^ivooq Kai e%e§pv)v,
Horn. Od. 13.330-32). Here Athena feels compelled to stick by Odysseus due to
their shared intellectual gifts (the key terms being noema and anchinoos). The

18. Her knowledge is also referred to at Hel. 317, 325-26, 530, 818-23.
20 THE ANCIENT QUARREL

goddess's capacity in this arena is prominent elsewhere, as when Hesiod attributes

superlative intelligence both to her mother, Metis, and to Athena herself (Th.
886-96); in addition, both he and Pindar mention the birth of Athena from the
head of Zeus (Th. 924 and Ol. 7.35-37, respectively).
As indicated above, a third category of etymologies comprises instances in
which onomata are connected with actions. One relevant subdivision is a type
of activity by which an individual or group is distinguished. The tragedians de-
vote much attention to the name "Apollo," and their analyses frequently under-
score the god's capacity for, even propensity toward, destructive action by link-
ing his onoma to the verb apollumi ("destroy"). In a famous passage from Aeschylus'
Agamemnon, for example, Cassandra bemoans her fate as follows: "Apollo,
Apollo! Lord of the ways, my destroyer. You have undone me once again, and
utterly" (omoXA,ov conoMov, dymch', coioM-cov e|i6c;- cmcoteocxc; yap o\) u6A,u;
TO 8emepov, 1080-82, with verbatim repetition of the first two lines at 1085-8
In Euripides' Orestes, Helen instructs Hermione to go to Clytemnestra's grave
and to "implore her to be gracious to us all, to my husband and me and these
poor children whom Apollo has destroyed (TOW T' dGAioiv ToivS', ovc, cxjtG&eoev
Geoq)" (119-21). Commenting subsequently on what he anticipates to be the fate
of Orestes and Electra, the messenger remarks that "neither your high birth nor
Apollo in his shrine at Delphi helped. No, Apollo has destroyed you both
(954-56).
Similarly, both Homer and Aeschylus associate Clytemnestra's name with the
pernicious course of action she undertakes. In the Odyssey, Agamemnon's shade
tells Odysseus how KXwai|ivf|OTpr| 5oA,6|iiyci<; ("wily Clytemnestra") slaughtered
Cassandra and recounts her disgraceful behavior toward him; he then notes that
"there is nothing more dread or more shameless than a woman who puts into
her heart such deeds, even as she too devised (e|jr|craTO) a monstrous thing,
contriving death for her wedded husband" (11.421-30; see also [ir\aaxo, 24.199).
Similarly, in Agamemnon Cassandra expresses concern about the destructive
intentions that Clytemnestra is harboring: "Ah, for shame, what can she pur-
pose now (icoTioTioi, ti Ttoxe p.f|8eT(xi,)? What is this new and huge stroke of atroc-
ity she plans (|J,r|8eTOH.) within the house to beat down the beloved beyond hope
of healing?" (1100-1103).
Both Aeschylus and Euripides highlight the descriptive content of the name
"Polynices." In Aeschylus' Seven against Thebes, Amphiaraus calls Polynices by
name, dwelling two times on the latter part (derived from neikos, which means
"strife"; 577-79). Some lines later, Eteocles observes that Polynices is quite fit-
tingly named, meaning that, true to his appellation (erttovu^cp Kcxpra, 658),
Polynices has turned out to be a cause of strife with reference to mortals and
their affairs. I9 Euripides offers the same basic treatment of Polynices' name when

19. Subsequently, Aeschylus extends the descriptive content of Polynices' name to cover
Eteocles as well: "They have earned their name too well and 'men of strife' they have perished
 LITERARY ETYMOLOGY 21

the poet has Eteocles speak to him as follows: "Leave this place; your name means
'quarrel' and our father named you well"
Phoen. 636-37). Subsequently,
Antigone addresses these words to her dead brother: "O Polynices, you followed
your quarreling name (e<|nx; dp' ETtcavuuxx;)" (1493), reinforcing the point with a
mention of "your strife which was more than strife" (1495).
In another relevant passage, Hesiod remarks that Hermes named a certain
woman "Pandora, because all they who dwelt on Olympus gave each a gift, a
plague to men who eat bread"
, Op. 81-82). Playing on the ad-
jective neon, Sophocles has Odysseus ask Neoptolemus if he would "do some
rash thing (11 veov) now" (Phil. 1229); on this interpretation, the name means
"Rash-warrior." In another noteworthy case, Aeschylus associates the name Persai
("Persians") with the verb pertho ("waste," "destroy"): Atossa reports that she has
often been haunted by dreams since her son Xerxes "gathering his host had gone,
his will to pillage (jtepoai) Greece" (Pers. 177-78). By using persai, the aorist
infinitive of pertho, Aeschylus makes the verb and name identical in spelling;
interestingly, Aeschylus employs the vocative Persai only seven lines earlier (171),
and the proximity of the two lines reinforces the etymology. Elsewhere, Hesiod
tells how the Titans received their name on account of the fact that "they strained
(raaivovTac;) and did presumptuously a fearful deed" (Th. 209-10).
Finally, Helen's name constituted a popular object of investigation. In a
memorable passage in Aeschylus' Agamemnon, one finds the Chorus musing as
follows about the striking match between the descriptive content of Helen's
onoma and the destructive outcome of a course of action involving her:

Who is he that named you so

appropriately in every way?
Could it be some mind unseen
in divination of your destiny
shaping to the lips that name
for the bride of spears and blood,
Helen, which is death? For fittingly
death of ships, death of men and cities
from the bower's soft curtained
and secluded luxury she sailed then,
driven on the giant west wind.

through impious intent" (o'i 8f\t' 6p9ax; Km entovuniav Kcd jioXuvevicei? coXovt' daepel 5iavota,
829-31).
22 THE ANCIENT QUARREL

In interpreting Helen's name, Euripides uses haired in the same sense as Aeschylus,
stating that she was the cause that Greece captured and destroyed (el^e) Troy
(Andr. io$-6).21
In the passage from Agamemnon involving Helen's onoma, the Chorus won-
ders who could have given a name so completely fitting. 22 The case of Helen,
while far from unique in raising the issue of appropriateness, is unusual insofar
as the father of the name's referent is none other than Zeus himself. Since Zeus
knows past, present, and future, it is fully within his power to assign his daughter
a name appropriate to the sort of individual she will become. One might argue,
however, that this passage's meaning is somewhat ambiguous since the Chorus
could have one of two things in mind: It might be asking, "Who in the world
gave her this name?" where no specific candidate is envisioned. Alternativel
its message might be, "Who gave her this name? None other than Zeus." In
assessing these possibilities, it is worth noting that the Chorus of Argive Elders
begins by asking who could have given a name so wholly true, then raises the
possibility that an invisible power is responsible for the assignment: specificall

20. With regard to the prevailing view of Helen's onoma, it is worth noting the remarks of
Gorgias in his Encomium of Helen: "The man who says rightly what ought to be said should also
refute those who blame Helen, a woman about whom both the belief of those who have listened
to poets and the message of her name, which has become a reminder of the calamities, have been
in unison and unanimity" (io\> 8' cruToi) dv8p6i; A,e!;ai te to Ssov 6p6ak; iced e^eyijai TOIJC;
ue|J.<|>ouevo'U[; 'EXevTiv, yuvaiKa Jiepi fj<; onoififflvoi; KCU 6uox|n)xo<; yeyovev fj te td>v TioiriTtov
aKOuacxvTCOv 7iicm<; r\ TE m\> ovonaTOt; <t>rinr|, 6 -cwv cn>|i<|>opa)v (ivii|iT| yeyovev, 2; the text and
translation are those of MacDowell 11982]).
21. In a related case involving a river, Aeschylus underscores the fierce character of its activ-
ity (PV 718-21), remarking that this body of water, called "Hybristes," "well deserves its name"
(TCOTajiov oii X|/eu5(6vDHOV, PV 717). Although a tie to hubris ("insolence") would normally lead
to an assignment's placement with those privileging bearers' attitudes and character traits, I lo-
cate the current passage here due to its focus on the potentially destructive character of the river's
activity, namely, the strength of its current. Regarding the present subcategory, cf. Eur. Ion 996-
97, where the aegis of Athena gets its appellation "from how she rushed into the battle" (To8'eaxev
6vop.a Gecov 61' rji^ev eg 86pi>); here an object closely associated with an individual gets its onoma
from the manner in which that individual initiates a course of action in which the entity has a
role.
22. In the ensuing discussion of this passage and of namers' supernatural inspiration, I am
indebted to the suggestions of Kenneth Dover.
 LITERARY ETYMOLOGY 23

the Chorus wonders whether someone unseen (UT| lie, ovtvv' oi)% opcbixev) gave
Helen this name based on that individual's foreknowledge of what was to be,
and thus achieved success.23 About the unseen namer Fraenkel (1950, 330)
remarks: "We do not know whence his influence comes; it seems a case of one
who operates tyaveic; dXaoTCOp f| KOKXX; Scdficov 7co0ev (Pers. 354). rcpovoica m\>
7ie7tptt>|ievot> belong only to a god or a daimon."24 What Fraenkel does not con-
sider is the possibility that Zeus himself— as Helen's male parent—is envisioned
by the poet as having assigned his daughter's onoma. I submit that Aeschylus'
phrasing lends at least some plausibility to the claim that the Chorus is pointing
to Zeus's role as namer.
What is of most fundamental interest regarding Agamemnon 681-92, how-
ever, is the reasoning process that one can trace in the Chorus's observations
about the naming of Helen. "Success" is said to be achieved in this endeavor
because the impact of Helen's actions has turned out to be precisely that pre-
dicted by the descriptive content of her onoma. In fact, according to the line of
reasoning reflected in the passage, it is because the assignment fits the adult bearer
so well that the Chorus is led to its initial musing about the name's source. That
this is so becomes evident in what follows the long opening question (681-87),
where the Chorus indicates its underlying motivation in making the inquiry at
all (687-92); the phrase ercei TipeTiovTCQc; ("For fittingly") is crucial because it
serves to link the latter portion of the Chorus's remarks closely to the former
and, more importantly, indicates a causal sequence moving from the second
sentence to the query made at the outset.25 It is to be expected that the Chorus
would focus on Helen's nature since the appropriateness of her name can only
be judged based on the sort of person she becomes. Because the assignment is
fitting, yet was made at birth, one is led inevitably to speculate about a super-
natural influence playing a role at this initial stage.
The case of Helen represents one of many in Greek literature in which the
appropriateness of an onoma given at the time of its referent's genesis can be
judged only at a later date, after the individual has reached maturity. The issue
of supernatural guidance might seem especially relevant in a case like hers, in
which the father is all-knowing Zeus, yet the reflective sequence observable in
the Chorus's remarks can be generalized to cover other instances in which fea-
tures and extensions of mature individuals match the descriptive content of names

23. Of course, if the name-giver is in fact invisible, this would place strong constraints on who
else besides Zeus could play this role. It is important to emphasize, as Fraenkel (1950, 329) does
in his commentary on Agamemnon, that \a\ is not used here to indicate that a negative response
is anticipated; rather, "this gives the wrong tone to the question in this passage . . . where the asker
of the question expects an affirmative answer or at least thinks it probable." Fraenkel also refers
here to Soph. OC 1500-1503, where "the initial question is developed" in just the same way.
24. Cf. Neustadt (1929, 246), whom Fraenkel cites in connection with this point.
25. Fraenkel (1950, 331) says very little about this particular instance of jipEJtovTGx; and does
not mention 8Jiei at all.
24 THE ANCIENT QUARREL

that were assigned at birth. In fact, once one raises the issue of divine provenance
in a single instance, it seems inevitable that one would wonder whether such a
factor were at work in naming more generally. Moreover, insofar as it is the ex-
istence of consonance between name and nature that leads to certain assump-
tions about an assignment's divine origin, the ontological status of the namer
(i.e., divine or mortal) is not of decisive importance. The idea that mortal par-
ents may be viewed as subject to supernatural influences is brought out clearly
in Euripides' Phoenician Women, where it is because the descriptive content of
the name "Polynices" meshes so well with its bearer's nature that Eteocles claims
Oedipus' assignment to stem from divine foreknowledge (6vop,a.. . Kemp eGeto
ooi 0eia itpovoia, 636-37).26
Another subcategory of assignments in the third class, which ties onomata to
actions, stresses an individual's or group's temporal relation to a particular course
of action or to others engaged in a certain endeavor. Such a tie is emphasized,
for example, when Homer presupposes the more standard interpretation of the
name "Neoptolemus," according to which it means "New-warrior": during his
Underworld visit, Odysseus informs Achilles that "in my own ship I brought him
out from Scyros to join the Achaeans under arms" (Od. 11.508-9) and goes on t
tell of Neoptolemus' value to them both in counsel and in battle (510-37).
Sophocles assumes the same analysis of this name, as when Philoctetes replies
as follows to Neoptolemus' comment that he is "at present bound from Troy":
"From Troy? You did not sail with us to Troy at first" (Phil. 245-47; see a^so 7°~
73 and 348-51); in fact, the state of affairs reflected in the descriptive content of
the name makes the play's action possible in the sense that it is only because
Neoptolemus is not a member of the original group of warriors that he has a
chance of forging a relationship with Philoctetes.27 In another instance, Athena
tells the sons of the seven warriors fighting against Thebes that throughout Hellas

26. Wyckoff s translation (1959) of lines 636-37 omits mention of the divine Jipovoia guid-
ing the assignment; in ray view, the idea of supernatural governance is crucial here, and its pres-
ence in the Greek should therefore be reflected in the translation.
27. For other cases in which names' semantic constitution is invested with a framing role,
one may consult Pindar's third and seventh Nemean odes. As concerns Nemean 3, which com-
memorates Aristoclides' victory in the pankration, Nisetich (rg8o, 239) notes that "the two com
ponents of the victor's name, Aristo-kleidas, signify 'superiority' and 'glory.' The second element
in the name (-kleidas) accounts for Pindar's identification of the Muse of this ode with Kleo ('Glo-
rifier'). . . . Superiority and glory apply even more impressively to the victor's homeland, which
quickly becomes the theme of the ode"; in the ode itself, see Pindar's remark that Aristoclides
has "bathed this island [i.e., Aegina] in the speech of renown (etncXei Jioyra)" (68). Regarding
Nemean 7, having observed that the victor and his father belong to the Euxenid clan, descen-
dants of Euxenus, Nisetich states (1980, 259) that "the theme of friendship between Pindar and
his patrons, and between his patrons and the gods, takes up a great deal of space" in the ode; he
notes subsequently (261) that "Nemean 7 might be considered a full-scale treatment of the poet's
role as the victor's friend or xenos." For relevant material in the ode itself, see lines 6-8, 6r, 64
65, 86-89. On the centrality of xenia to Pindar's odes more generally, see Nisetich 1980, 46.
 LITERARY ETYMOLOGY 25

they will be called the "Epigoni"; this group receives its onoma based on its tem-
poral relation to another collection of individuals, namely, their fathers, whose
deaths they are to avenge upon reaching manhood (Eur. Supp. 1224-25).28
A third subdivision of the present category consists of instances in which
authors' focus is on the role a place or individual has in someone's plan of ac-
tion. One finds an interesting case of this type in Euripides' Helen, where Helen's
name is linked implicitly with the verb haired in the sense of "capture." At the
close of that play, Castor tells her that where Hermes "first defined your place
when he caught you up from Sparta . . . , stealing (KAEXJ/OK;) you . . . , where the
island stretches to guard Acte, shall your name be known as Helen ('EA.evr|),
meaning Captive, for mankind hereafter; because you were stolen from your
house" (1670-75). Instead of depicting Helen as a destructive influence, as he
does in Andromache, Euripides stresses her role as an object of coercion. In
another pertinent instance of this type, Herodotus states that a certain place re-
ceived the name "Aphetae" because the Argonauts intended to make it their
point of departure after drawing water for the ship (ev9eihev yap ejieMiov
ij5p£U0dn,£voi eq TO nekayoq d(|)fioetv) (7.193.2). The source of this assignment
is unspecified (Herodotus says simply that ETU TOWOD TCO xcopcp owofxa yeyove
'A<j>£TCu); nevertheless, the place comes to have this name based on the role it
played as a launching point for the Argonauts in the context of their mission.
A fourth category includes cases in which the descriptive content of onomata
stresses significant effects that their referents have on mortals. Thus, Apollo's ca-
pacity to induce fear in mortals is reflected in Aeschylus' derivation of the epithet
"Phoebus" when Atossa, mother of Xerxes, says that "then to Phoebus' hearth I
saw an eagle fleeing: dumb in dread I stood" (6pd> 5e fyeiiyovx OIIETOV npoc,
eo^dtpav <3>oi.poi>, (|>6pcp 5' afyQoyyoc, ecrca6T|v, Pers. 205-6). Furthermore, in a
memorable passage Homer speaks of the impact that certain gates made of horn
or ivory (KEpaEoai, eXe^avii) have on human beings and their concerns: dreams
coming through the latter gate deceive (eXe^alpovtai), while those passing through
the gate of horn "bring true issues to pass" (ETU^CI Kpcdvoi)cn) (Od. 19.562-67).
A fifth classification includes names that are tied to entities with which an
individual or group is associated. These might be objects used in the performance
of actions, as evidenced by Homer's observation that King Areithous received
the additional name "Club-wielder" (TOY E7iiKA,T|oiv KOpt>vr|Tnv) because he
fought, not with a bow or spear, but with an iron club (ai8r)p£iT| Kopwrj)
(II. 7.138-41). In addition, the poet ties the name "Penelope" to TITJVTI ("woof)
when he has her deceive the suitors by allegedly spinning a shroud for Laertes
in the event of his death (Homer presents the tale at Od. 2.93—110; see also 19.137—
56 and 24.128—48).Z9 An assignment in this fifth category might also be tied to an

28. Pindar also uses the term 'Eiriyovoi (Pyth. 8.42).

29. According to West (Heubeck, West, and Hainsworth 1988, loj), it is possible that the story
about Penelope's spinning activity arose from an imagined derivation of her name from
26 THE ANCIENT QUARREL

article or type of clothing that an individual or group is said to wear, as when

Homer tells how at dawn the nymph Calypso clothed herself and placed a veil
(KaA,TJ7rcpriv) on her head (Od. 5.228-32).30 In a second case falling in this sub
category, Herodotus mentions a certain people named for the black garments
favored by its constituent individuals: "The Melanchlaeni all wear black cloaks —
hence their name"
4.107).
Sixth, an onoma may be assigned based on a characteristic of the place from
which its bearer hails, as when Athena appears to Telemachus as "the son of
wise Anchialus ('Ay^idAmo . .. i)ioc;) . . . lord over the oar-loving Taphians"
(Horn. Od. 1.180-81); the father's name is formed from dy%iaA,oc;, which means
"near the sea" in the case of cities and "sea-girt" in that of islands.31 Alternatively,
in a seventh type of case, the onoma of a natural, inanimate entity is traced to an
important individual with whom it is linked. Illustrative of this approach is
Prometheus' foretelling that lo will eventually pass through the channel of Maeotis,
"and hereafter for all time mortals shall talk about your crossing, and they shall
call the place for you Cow's-ford" (eotoa 8s 0vr|Tol<; eioaei Kojoc, \ieyac, Tr\<;
of|<; Tiopeiaq, BoaJiopoc; 8' eiccovunoq KeKXrioeiai, Aesch. PV 732-34).
As noted, an eighth category of analyses stresses the role an individual plays
in a particular social or familial context. In a prominent instance of this type,
both Homer and Euripides draw attention to the descriptive content of the names
"Hector" and "Astyanax," as well as to the semantic ties between them. Homer's
etymology of the former is most explicit in Andromache's use of the verb echo
("protect") when addressing her dead husband: "I think that our son will never
come of age, for before then head to heel this city will be sacked, for you, its
defender, are gone, you who guarded the city, and the grave wives, and the in-
nocent children (rj yap oKatXac, ETUOKOTTOC;, oc, TE jiiv awfjv pwKei), e%ec; 8'
dX6%oiJc; KeSvcn; Kai vfptia xeKva)" (II. 24.727-30). Hector's role as protector,
reflected in his name, is also stressed in Priam's reference to Achilles' killing a
few days earlier of his sole remaining son, "who guarded my city and people"
(eipmo. .. doit) Kai amove,, 24.499). Elsewhere, Andromache speaks of Astyanax,
"whom the Trojans have called lord of the city, since it was you [i.e., Hector] alone
who defended the gates and the long walls" (6v Tpcoeq emKA,T\ovv KaXeoww
0101; yap ocfav epnoo nvhac, Kai xei%ea jiaKpd, 22.506-7); emphasizing the same
semantic and causal relationships, Homer says that while Hector called his son
"Scamandrius," other people called him "Astyanax," "because Hector alone

6>.6jtT(Q ("tear out") or ^.enco ("strip off). On the name's connection to 7iT|VTi and ^COTIT| ("cover-
ing," "robe"), see Peradotto (1990, 107-8), who suggests that Penelope's onoma is "designed for
the heroine of just such a story."
30. Cf. her association elsewhere in the poem with the notion of concealment: Calypso is
said to have kept Odysseus back in her hollow caves (1.14-15), and Hermes' remarks emphasize
how isolated her island is (5.99-102).
31. For these meanings of ayxta^oi;, see LSJ.
 LITERARY ETYMOLOGY 27

protected Troy" (oloc, jap eptteto "IXiov "EKTCOp) (6.402-3).32 In Euripides'

Trojan Women, Andromache addresses Hector, now dead, as her defender (oa^
5d(j,apTOc; a^icap, 590). Moreover, Hecuba laments the fact that Astyanax was
killed at such a young age when he might instead have reached adulthood and
died fighting on behalf of his city, that is, when he might have performed the
same civic function as his father (1168); shortly thereafter, the Chorus addresses
Astyanax as "you who were once a great lord in my city" ((isyaq E|ioi TIOT' (Sv
dvaKtcop 7EO/leco<;, 1217). While the Chorus's direct reference is to Astyanax's
former high status in Troy (pot' on), the noun anaktor ("lord") cannot but link
Astyanax with Hector by alluding to that protective role common potentially to
father and son and encapsulated in both onomata.
One might conjecture that where the descriptive content of proper names
is positive, namers' hopes combine with the magical idea of the onoma as
nature-conferring to produce certain expectations for performance (examples
of names given based on parental hopes include "Hector," "Astyanax," and
"Theoclymenus"); in this sense, the name would provide the individual with a
model for imitation. 35 It is only after individuals reach maturity that one can
judge whether the relevant developments have occurred. Insofar as hopes and
expectations—as reflected in names' descriptive content—are met, the view that
the parents in question are supernaturally inspired finds support.34 If one thinks
of names, at least in part, as prophecies, then the idea is that one must wait some
time before the truth or falsehood of their predictions can be established. At first
glance, there might seem to be two incompatible ideas at work here: that of pre-
diction based on supernatural awareness of how the adult bearer will turn out
and that of the name's helping to shape the outcome in question. In fact, how-
ever, the clash is only apparent. The existence of individual natures, a belief in
whose salience is prominent in literature, places constraints on capacities and
achievements in particular cases. The idea is that successful namers are some-
how able to anticipate what is possible for individuals in the cases of naming
over which they exercise control and to aid in these individuals' development
by singling out relevant features of them in the onomata that they assign.35

32. My translation.
33. With respect to the line of reasoning developed in this paragraph, I am grateful for the
suggestions of Kenneth Dover. Regarding "Theoclymenus," see Eur. Hel. 8-10; this name was
assigned to the son of Proteus and Psamathe based not only on the father's piety but also on the
parents' hope that their son, once mature, would display that same attitude toward the gods. On
the general topic, cf. the observation of Higbie (1995, 189) in her study of Homer that "the name
given by characters to a child expresses the nature and qualities that the child is hoped to inherit
from a parent, usually the father."
34. On the topic of supernatural influences on namers, see the earlier comments occasioned
by Aesch. Ag. 681-92.
35. For an emphasis on the presence and salience of individual natures (referred to by phusis
and phuein), see, e.g., Soph. A;'. 549, OT 674, Phil. 79, 874, 902, r3ro; Eur. Med. 1343-45, ZA 55
930, 1411, Hel. 1003, Ion 240, 643, Cyc. 649, Hipp. 79, Phoen. 395, Bacch. 315, Or. 126.
28 THE ANCIENT QUARREL

In another case of a name describing an individual's social function, Herodotu

mentions the Delphic oracle's instructing the Cyrenaeans to bring in someone
from Mantinea to set things right for them; in answer to their request for an
appropriate person, the Mantineans sent a man by the name of "Demonax"
("Leader-of-the-people") (4.161.2). A belief in the existence of deep connections
between onomata and reality governs instances, such as this one, in which a
name's descriptive content provides the basis for its bearer's involvement in a
particular course of action. A later example from Herodotus (9.90-92) illustrates
more vividly still the idea that names, due to their semantic constitution, may
impact reality itself: Three Samians approached the Greek commanders with a
message. One of their number, Hegesistratus (whose name means "Leader-of-
the-army"), appealed to the commanders to rescue the lonians from slavery.
Leotychides asked his name and, upon discovering it, required the man to sai
with the Greek fleet since Leotychides believed that the name would be a good
omen (oicovov TO owo^cx 7toiet)(J.evoq, 9.92.2). An onoma may, in addition, be
used to refer to an individual's anticipated future social role, as when Herodotus
mentions (4.155.2-3) a man who was renamed "Battus," battos being the Libyan
word for basileus ("king"), and claims that the priestess at Delphi addressed him
by this name because she knew that he would become a Libyan king.
Regarding the idea that a name's analysis may underscore the familial role of
its referent, there are passages in Oedipus at Co/onus which, if combined, sug-
gest that Sophocles takes Antigone's name to mean "Instead-of/Like-a-son." The
term gonai is used for "sons" (1192). Subsequently, Polynices, addressing Oedi-
pus, refers to himself as "your son —or if I am not truly your son . . . at least I am
called (KaA,OTJU£VO(;) your son," and uses the preposition anti with reference to
Antigone and Ismene (1323-26).36 Oedipus proceeds to inform Polynices that
Antigone and Ismene "have saved me, they are my support, and are not girls,
but men, in faithfulness. As for you two [Polynices and Eteocles], you are no
sons of mine!" (a'iSe p.' EKaro^oticnv, cuS' eiiai Tpoc|>oi, aiS' av8pe<;, o\) yuvaiicec;,
eq to ov)|a,7iovetv iJiieii; 8' COT' dlXov KOIJK eiaov) Tce^'UKaiov, 1367-69). Sophocles'
application of the descriptive content of the name of one of two siblings to both
individuals parallels Aeschylus' treatment of Polynices' name in Seven against
Thebes, where, as we saw, he uses the descriptive content of that onoma to cover
both the bearer and his brother, Eteocles.
Finally, onomata are tied to functions and special powers of the divine:'7 here
a deity's name may be traced to an element over which its referent has control,
to a special capacity of the individual, or to the pivotal role assumed by the ref-
erent on the cosmic plane. With regard to the first subcategory, Homer says that

36. One finds anti in the sense of "like" or "as good as" already in Homer, e.g., in II. 9.116-17:
dvtl vu jroU-div tarav e<mv dvip ov te Zeui; Kfjpi (juXTJon. See also Od. 1.70, 2.17, 3.414, 8.546
37. In general, I reserve this category for functions and powers that mortals do not have, and
passages that depict gods as having capacities that also belong to mortals, or doing things that
mortals too can do, are placed in other categories.
 LITERARY ETYMOLOGY 29

Hera shed thick mist (aer) around her horses (II. 5.775-76); reinforcing this ety-
mology, he associates the goddess with air again later in the work (14.282). In
what may well be the same etymological move extrapolated one step further,
Euripides has Helen say that Hera "made void the love that might have been for
Paris and me and gave him, not me, but in my likeness fashioning a breathing
image out of the sky's air (6|j,oi<aaci(T' eiioi e'i8coXov ei^ivow oupavoiJ c;w0eio'
GOTO), bestowed this on King Priam's son" (Hel. 32-35; cf. his connection of Hera
with aither at 241-51).
On several occasions authors stress noteworthy capacities of individual deities.
A special ability to avoid detection is at issue in Homer's treatment of Hades as
the "Unseen-one": Athena put on the cap of Hades, "that stark Ares might not
discern her" (|ifi|iivi,8oi6|3pi!J,o<;"ApT)<;) (II. 5.844-45). Hoping to direct Apollo's
tremendous power in a positive direction, and toward that end proposing the
name's connection to the verb luo ("release"), Jocasta addresses him as follows:
"I came as suppliant to you, Lycaean Apollo. . . . Grant us escape free (A,ixn.v
TIV') of the curse" (Soph. OT 919-21). Elsewhere, a Euripidean Chorus asks
Apollo to be Admetus' "redeemer from death" (XuTipioc; EK Gavdrcoi}, Ale. 224).
Aeschylus traces the name "Ares" to the noun am ("curse"): "A bitter and evil
divider of possessions, Ares, who made their father's curse a thing of utter truth"
(TtiKpoq 8' 6 xprpdTcov KOIKOI; 8dTT|Td<;"Apr|<;, dpdtv nmpcpciv uSeii; dA,a0f|, Sept.
944—46).38 In a passage concerned with the capacity of one god to bring evil to
another, Aeschylus (PV 910-11) associates Cronus' name with the verb kraino ("ac-
complish") when he has Prometheus say that "so shall at last the final consumma-
tion be brought about of Father Cronus' curse" (natpoc, 8' dpd Kpovou TOT' fj8r)
TtavTEXfix; KpavQriaETai) unless Zeus seeks and obtains Prometheus' aid.39
As mentioned above, the third subdivision of this category consists of assign-
ments that stress the instrumental role that a divine individual or group plays in
the cosmos. Notable in this regard is literary authors' frequent consideration of
the two forms of Zeus's name, Dis and Zeus. Hesiod's invocation of the Muses
at the outset of Works and Days links the form Dis (an old nominative for Zeus,
found in the oblique cases) to the preposition dia ("through," "on account of):
"Muses. . . tell of Zeus (Ai') your father and chant his praise. Through him (6v
TE 8id) mortal men are famed or unfamed, sung or unsung alike, as great Zeus

38. In a passage with a very different emphasis, Hephaestus bemoans the fact that Aphrodite
scorns him due to his physical deformity, claiming that she prefers Ares due to his attractiveness
and soundness or swiftness of limb (ouve%' 6 Kakdc, te Kai dpiiTtOQ) (Horn. Od. 8.308-11). Al-
though an etymological tie between Ares and artipos is conceivable, it is more likely that the
parallels here are on the phonetic plane.
39. For a different view, see Wecklein (1981, ad loc.), who treats the connection between Kronos
and kraino as phonetic ("alliteration of Kp-") rather than etymological. Griffith (1983, ad loc.)
does not note any sort of link, stating that "this 'father's curse' receives no further mention in the
play; nor do we hear of it anywhere else in ancient literature. If this is the first that the audience
has heard of it, then it seems rather a casual and pointless mention." That an etymological tie is
being forged here was recognized by Kranz (1933, 289).
30 THE ANCIENT QUARREL

wills (Aioq jj.eydA.oio EKT|TI)" (1-4). Aeschylus too connects this form of the god's
name with did when he has the Chorus in Agamemnon identify Zeus as the
ultimate source of the misfortunes of the House of Atreus: they happened "all
through Zeus, Zeus, first cause, prime mover. For what thing is done by mortals
without Zeus?" (5iod Aioc; Ttavamoi) TtavepyeTcr TI yap Ppoiolg dvei> Aioc,
lekeliai;, Ag. 1485-87). In Isthmian 3, in turn, Pindar ties the form Zeus to zen
("to live"): "O Zeus (Zeu), mortal man's prowess springs from you. His happi-
ness abides (£coei) longer when he reveres you" (4-5). Aeschylus underscores
the same link in a reference to what "is in very truth the seed of life-giving Zeus"
((trucn.£6o\>yevoc;... Zr[v6c, ecmv aK^Q&c,, Supp. 584-85).40 Moreover, Euripides
too stresses this tie when he has Apollo say that "Helen lives (^T\V), for being born
of Zeus (Zrivoc; yap ouoav), she could not die" (Or. 1635); here a direct causal
connection is clearly indicated.41
In another pertinent case, the Chorus in Euripides' Phoenician Women
mentions "the goddesses named together. . . Persephone and dear Demeter, who
is ruler of all, and is Earth, nurse of all" (Sicovunoi 9eod, Il£p<j£(]>aaoa Kai §iKa
Aa(a.aTrip 9ed, Tidvicov dvaooa, Tidvirav 8e Fa tpoc|>6c;, 683-86).42 This remark
indicates the derivation "Mother-Earth." Making the same proposal in the
Bacchae, Tiresias notes that Demeter and Earth are said to be identical: "Man-
kind . . . possesses two supreme blessings. First of these is the goddess Demeter,
or Earth—whichever name you choose to call her by (Arpr|Tr|p 0ed—yfj 5'ecrav,
6vo|j,a 8' orcoiepov pox)A,T) KdXei)" (274-76). Elsewhere, Euripides refers to
Demeter as "the Mountain Mother of all the gods" (He/. 1301-2; cf. 1320,1340,
1356), observing that while she grieved over her lost daughter, "earth, green gone
from her fields, would give food no more to mortals in the sown lands" (1327-
28). Finally, according to Herodotus, the Pelasgians "called the gods by the Greek
word theoi — 'disposers'—because they had 'disposed' and arranged everything
in due order, and assigned each thing to its proper division (KOOUXO 0evie<; id
jtdvTot 7ipf|y(j,cna Kai ndoaq vo^ai; ei%ov)" (2.52.1).
It is evident from the tremendous number and diversity of passages in which
authors offer etymologies that they were quite concerned with the issue of connec-
tions between elements of language and those of reality; most often they concen-
trate on showing how proper names, once analyzed, reveal something important
about their individual bearers' natures.43 Literary writers employ a wide range of

40. Tr. Smyth (1922).

41. Finally, regarding Zeus's onoma, one may consult Aesch. Ag. 160-75, with the interpreta-
tion of Quincey 1965, 147-48.
42. My translation.
43. The literary tradition does not have a technical philosophical notion of individual na-
tures according to which one seeks and employs a rigid set of necessary and sufficient conditions
for making identifications. Authors operate instead with a loose, nontechnical notion according
to which a particular individual, either mortal or divine, is widely recognized by a salient charac-
teristic, power, or type of activity.
 LITERARY ETYMOLOGY 31

criteria in their speculations about the descriptive content of proper names and
do not demonstrate an interest in being consistent or systematic: although certain
standards, for instance, that criterion involving bearers' attitudes and character traits
are especially popular, authors do not regularly adopt a particular approach. In-
stead they proceed haphazardly, based on the requirements of a particular con
text.44 Moreover, and quite importantly from Plato's perspective, they neither call
the value of the approach itself into question nor express doubts about crucial
assumptions underlying the enterprise: namely, that there are substantive connec-
tions between onomata and entities that etymological analysis reveals, and that
natures belong first and foremost to individuals. In the Cratylus, as we will see,
Plato employs the literary tradition's techniques and assumptions, with the ultimate
goal of discrediting them. Neither there, nor elsewhere, does he espouse the view
that one may achieve genuine insight by analyzing the constitution of onomata.

Appendix: Other Expressions of a

Belief in Substantial Ties between
Onomata and Elements of Reality
A belief in substantial connections between elements of language and of reality
is evident in arenas besides that of etymology. In this appendix, I comment on
the issue of changes in onomata for apotropaic purposes, the omission of onomata
due to a fear of (or for) their referents, and the use of curse tablets (defixiones) as
it bears on the theme of ties between names and their bearers. Greek literary
texts illustrate a strong belief that substituting an onoma with positive connota-
tions or descriptive content for one whose referent generates fear or superstition
could help to counteract its power. This view is strikingly evident in authors'
attitude toward and naming of what is on the left.45
In Greek literature, greater value and higher status is regularly assigned to what
is located on the right. Oaths, for example, are sealed with the right hand, as when

44. Not only do different writers provide divergent etymologies of the same name, as in the
case of Aphrodite (Hes. Th. 197-98; Eur. Tro. 989-90), but an individual author may himself off
distinct etymological analyses of a single onoma, as happens in Pindar's treatment of the name
"lamus" (Ol. 6.45-47, 54~5?)- 'n addition, a writer may offer multiple appellations for the same
individual, providing an etymological treatment of each, as Hesiod does in the case of Aphrodite
(Th.' 197-98, 199).
45. Among Presocratic philosophers, a distinct preference for what is on the right is evident
in Parmenides, as when the goddess greets him after his metaphorical transition from Night to
Day by grasping his right hand (DK 1.22-25). Moreover, Parmenides' opposed valuations of right
and left—positive and negative, respectively—are present in his embryology, according to which
"on the right boys, on the left girls" (DK 17). Aristotle (GA 763b3i~764ai) mentions Anaxagora
as among those who connect right with male and left with female in the process of generation.
Concerning the favoring of right over left, one may also consult the Pythagorean Table of Oppo-
sites, which is preserved by Aristotle (Metaph. 986322-26).
32 THE ANCIENT QUARREL

the nurse in Euripides' Medea says that "poor Medea is slighted, and cries aloud
on the vows (opKOtx;) they made to each other, the right hands clasped in eternal
promise (dvciKa?iev 8e 5e^ia<; Tticrav iieyioiriv)" (20-22); later in the play, in
fact, Medea reminds Jason of that earlier pledge—clinched with the right hand—
which he has now broken (495-9y).46 In addition, feelings and attitudes, promi-
nent among them affection and respect, are conveyed by touches of the right hand.
Thus, Medea tells her children to welcome Jason by grasping his right hand (Med.
899) and later asks for their hands so that she might kiss them (1070). Elsewhere,
Oedipus asks Theseus whether he might grasp his right hand as a sign of respect
and gratitude for Theseus' kindness toward him (Soph. OC1130).47 Moreover, the
right hand does the most central work in acts with religious import, as when Hecuba
carries a golden goblet containing wine in her right hand in order that a drink-
offering might be poured as Priam and the herald prepare to depart on the embassy
to Achilles (Horn. II. 24.284-85).48 In war, in turn, being on the right wing of the
army is considered to be more prestigious. This view is made explicit by Herodotus,
who says that Callimachus led the right wing of the Athenian army at Marathon
since it was customary in Athens at this time to bestow that honor on the War
Archon (6.m.i).49 In series, moreover, movement from left to right is preferred.

46. Elsewhere, Menelaus indicates his acceptance of a mutual oath (ciyvov opKOv) between
himself and Helen by instructing her to take his right hand (Eur. Hel. 835-38), and Heracles insists
that Hyllus' promise to him be confirmed in the same way (Soph. Track. 1181). For other relevant
passages, see Horn. II. 2.339-41, 4.155-59. Cf. Orestes' request to Pylades to seal his promise to
build his tomb with the clasping of right hands (Eur. JT 701-2). In addition, pleas are made, at
least in part, by touching the right hand of those whom one entreats, as when Iphigenia requests
that the temple maidens demonstrate their loyalty by not revealing what they know (Tipoc, oe Se^id?
ae iced a' i.KVO'OUca, IT 1068-69). See also Eur. IA 909-10, where Clytemnestra entreats Achilles
to preserve his own reputation by helping to save Iphigenia from death: "I beg you, by your beard,
your right hand (npoi; oe Se^icig), and by your mother's name—cleanse your own name of this
reproach." Regarding this type of case, cf. Eur. Hec. 342. For votes registered by the raising of the
right hand, see Aesch. Supp. 607.
47. With respect to feelings and attitudes, one may also consult Eur. IA 679, 866; Horn. II.
10.542; cf. Horn. Od. 1.121, where a stranger's right hand is grasped as a sign of welcome. Penelope
recalls Odysseus' final words to her before his departure, which were preceded by his grasping of
her right hand and wrist (Od. 18.258). Here the touch seems to be above all a gesture of affection,
but on some level it also conveys a recognition of Penelope's sadness. One person may grasp
another's hand for the primary purpose of acknowledging painful feelings and emotions that the
latter is experiencing, where the touch aims to diminish the feeling or emotion in question. In
the Iliad, for example, Achilles takes Priam by the right hand at the wrist so that he will not be
fearful (24.672). Elsewhere, Alcestis' handmaid notes that "all the servants in the house were cry-
ing now in sorrow for their mistress. Then she gave her hand to each, and each one took it (f| 8e
6ef;idv 7tpov>teiv' eKdota))" (Eur. Ale. 192-94).
48. For the right hand's salience in the context of animal sacrifice, see Eur. El. 812; Hel. 1581.
On the importance of the right-hand side in connection with the dedication of grain, see Eur. IA
1472. Elsewhere, Dionysus notes that real Bacchantes hold the wand in the right hand and raise
it as they raise the right foot (Eur. Bacch. 941—44). Concerning the religious arena, cf. Hdt. 1.51.1,
where the bowl of greater value is placed on the right side of the entrance to the temple at Delphi.
49. Regarding the organization of the Greek army at Plataea, see Hdt. 9.26.6-28.6.
 LITERARY ETYMOLOGY 33

This preference is demonstrated in the passing of a range of objects, among them

wine cups (Eur. Rhes. 364) and lots, as when the lot of Aias is shown to other warriors
(Horn. II. 7.184).50 Even the gods are said to observe this sequence (see II. 1.597,
where Hephaestus pours drinks for the other gods, moving from left to right).
In addition, bird signs are regularly interpreted as positive if the flight of the
bird or birds is on or to the right, as when the heron Athena sends to Diomedes
and Odysseus appears on the right near the path that they are taking (II. 10.274).
Later in the Iliad, Hecuba exhorts Priam to pray to Zeus "for a bird of omen . . .
one seen to the right (Se^iov), so that once your eyes have rested upon him you
can trust in him and go to the ships of the fast-mounted Danaans" (24.292-95);
all then rejoiced when, following Priam's request (310-13), a dark eagle "swept
through the city appearing on the right hand" (319-20). In the Odyssey, Zeus
sends a sign to Telemachus —who has just appealed to the god in a rebuke of
the suitors —in the form of a pair of eagles which "dropped on the heads of the
crowd . . . wielding their talons, tearing cheeks and throats; then veered away on
the right hand through the city" (2.152-54). Telemachus receives another such
sign of support later on, as he takes leave of Menelaus: "Even as he spoke, a beat
of wings went skyward off to the right (5d;i6<; opviq) —a mountain eagle, grap-
pling a white goose in his talons, heavy prey hooked from a farmyard. Women
and men-at-arms made hubbub, running up, as he flew over, but then he wheeled
hard right (syyuOev Sd;to<;) before the horses—a sight that made the whole crowd
cheer, with hearts lifting in joy" (15.160-65). For yet a third time, Telemachus
receives divine support in the form of a positive bird sign "when a hawk, Apollo's
courier, flew up on the right" (15.525—26).51

50. In another pertinent case, Odysseus, following Athena's instructions, "appealed to the
suitors, one after another, going from left to right (evSe^ia), with open palm, as though his life-
time had been spent in beggary" (Horn. Od. 17.365-66). Antinous' subsequent instructions to the
other suitors regarding attempts to string the bow of Odysseus incorporate the same preference
(Od. 21.141-42).
51. Theoclymenus' interpretation underscores the sign's positive import: "A god spoke in this
bird sign on the right. I knew it when I saw the hawk fly over us. There is no kinglier house than
yours, Telemachus, here in the realm of Ithaca. Your family will be in power forever" (15.531-
34). For other relevant passages, see Horn. II. 13.821-23; Od. 24.311-13. Regarding divine omens,
one may also consult II. 2.353, where Nestor says that Zeus "flashed lightning on our right as a
sign of favor"; II. 9.236-37; Eur. Phoen. 1189-91.
Furthermore, positive descriptions of mental and physical gifts contain dexiotes or dexios (on
its own or in a compound), as when Herodotus mentions a wreath that "was granted to Themistocles
. . . for his ability and skill (aotyir\q . . . Kai 5eJ;i6TiTK><;)" (8.124.2). In the ninth Olympian ode,
Pindar notes that the gods gave Epharmostus "sure hands, lithe limbs (Seijioymov), the face and
brow of power" (no—11). In Isthmian 5, Pindar says that "among pancratiasts, I praise Pytheas for
having steered his brother's blows aright—he's clever with his hands (xepoi 8e^iov) and a match
in wits" (59-61). Elsewhere, Herodotus states that "the Greeks have never been simpletons; for
centuries past, they have been distinguished from other nations by superior wits" (ejtei ye cmeKpi6r|
EK 7iaA.avtepou toi) pap Papon £9veo<; to 'EA^TIVIKOV eov KOI de^itOTepov Kai ei>T)6ir|<; TiA.iBio'u
a7cr|A.A.ayiievov M,aA.A,ov, 1.60.3).
34 THE ANCIENT QUARREL

The foregoing discussion provides abundant evidence that the Greeks had a
marked preference for what is located on the right or moves in that direction.
Strongly implied in such passages is a negative judgment, relatively speaking, of
what is on the left, and in fact writers make this assessment explicit. While, as
we saw, bird signs on the right are viewed as auspicious, those on the left are
regarded with trepidation. One notable example occurs in connection with the
Trojans' attempt to set fire to the Greek ships: as the Trojans "were urgent to
cross [the ditch], a bird sign had appeared to them, an eagle, flying high and
holding to the left of the people (aietoq \>\|raceTr|<; en apioiEpa Xaov eepycov)
and carrying in its talons a gigantic snake. . . . And the Trojans shivered with
fear as they looked on the lithe snake lying in their midst, a portent of Zeus of
the aegis" (Horn. II. 12.200-209). Polydamas interprets the omen as portending
a negative outcome to their project and enjoins Hector to desist (12.215-29; see
especially his repetition of the fact of the eagle's "flying high and holding to the
left [ETC' dpioTepd] of the people," 219).52 In the Odyssey, in turn, the suitors were
plotting the death of Telemachus, "when from the left an eagle crossed high
with a rockdove in his claws (avrap 6 TOIOIV dpiotepoq r\k\>8ev opviq, cdsTOq
•u\|n.7teTr|<;, e%e 8e tpipcflva TieAsiav). Amphinomus got up. Said he, cutting them
short: 'Friends, no luck lies in that plan for us, no luck, knifing the lad. Let's
think of feasting'" (20.242-46)."
In order to counteract any negative effects associated with something's posi-
tion on the left, writers may speak euphemistically, substituting euonumos ("well-
named" or "of good omen") for an'steros.54 In a move that appears to have an
apotropaic aim, Prometheus says, "The flights of crook-taloned birds I distin-
guished clearly—which ones are auspicious by nature, and which sinister"
(yaiavi/covuxcov TE nifioiv oicovrav OKeepmt; 5icbpio', omvec; IE 8e^ioi ^-uaiv
etKavujioix; xe, Aesch. PV488~9o).55 In another context, Antigone speaks of her

52. In his reply to Polydamas, Hector refuses explicitly to regulate his conduct according to
the behavior of birds (12.237-40), but his failure to heed the bird sign that appeared on the left
does not have a positive outcome.
53. Cf. Herodotus' discussion (2.30.1-2) of the "Deserters—a people whose name is Asmach,
a word which would mean in Greek 'those who stand on the left hand (ei; dpioTepf|i; x^ipoi;) of
the king.' They were a body of men two hundred and forty thousand strong, of the Egyptian warrior
class, who went over to the Ethiopians during the reign of Psammetichus."
54. Where the euphemistic replacement of terms is not at issue, euonumos is used in a range
of positive descriptions, especially of individuals (Hes. Th. 409; Find. Ol. 2.7, Pyth. 11.58, Nem.
8.47) and cities (Find. Nem. 4.19, 7.85,11.20). The adjective is employed with reference to justice
at Find. Nem. 7.48.
55. Tr. Smyth (1922), with modifications. Grene ("which of them were in nature prosperous
and lucky" [Benardete and Grene 1991]; "which of them were propitious or lucky by nature"
[Lattimore, Benardete, and Grene 1956]) treats dexios and euonumos as synonymous. It is doubt-
less true that for the Greeks what is dexios is euonumos. However, the contrast between left and
right is part of normal Greek usage and likely to be in play here, especially given the context in
which the terms are used.
 LITERARY ETYMOLOGY 35

brothers as "stricken through the left sides (?>\ e\)cov6p,cov Teruunevoi), stricken
indeed, through sides born of a common mother" (Aesch. Sept. 887-90). Else-
where in Aeschylus, Danaus instructs his daughters to "prepare quickly white
suppliant wreaths, sign of Zeus sacred, held in the left hand (e^ouaca 8ia xeprav
ewovuiKQv)" (Supp. 191-93). In Sophocles' Women ofTmchis, in turn, the nurse
tells how Deianira prepared to kill herself by uncovering "her whole side and
her left arm" (Tt^etipdv craaoav (oXevrrv T' e\)c6vu|J,ov, 926). Herodotus uses
euonumos with kerns on several occasions when speaking of the left wing of an
army (see 6.111.1,9.28.6,9.46.3,9.47); this euphemistic employment of euonumos
is not surprising given the belief, mentioned above, that a position on the right
wing was viewed as more favorable.
A shift in onoma for apotropaic purposes also occurs in the case of the
Eumenides. The Furies, as depicted by the tragedians, evoked great fear in
mortals, and this fear is evident in mortals' approach to the naming of these
deities.56 In Sophocles' Oedipus at Co/onus, Oedipus is told, regarding the spot
where he sits, that "most dreadful are its divinities, most feared, daughters of
darkness and mysterious earth" (cd jap e[i§o$oi 0ecd o(()' e%o\ioi, Ff)<; xe tccd
ZKOTOU Kopai, 39-40). Having asked under what name he should invoke them,
he is informed that the people of Colonus prefer to address them as "Gentle All-
seeing Ones" (xac, navQ' oprooac; E"i)|a.evi5a<;, 42). The mortals in question assign
them the name "Eumenides" based on their fear, that is to say, based on the
attitude that they actually want these divinities to have toward them. This come
out clearly when the Chorus asks the deities, in effect, to be true to this assigned
name with positive descriptive content: it tells Oedipus to repeat the prayer that
"as we call them Eumenides, which means the gentle of heart, may they accept
with gentleness (EC; eiJ^evcov) the suppliant and his wish" (486-87).57
In the cases discussed in what precedes, onomata whose referents are viewed
as productive of fear may be changed for apotropaic purposes. One may also re-
frain from uttering certain onomata on the same or related grounds. Oedipus at
Colonus illustrates clearly the point that fear of entities' power to harm can lead to
the conscious suppression of a name: the Chorus speaks of the Furies as "those
whom it's futile to fight, those whom we tremble to name" (127-29). The same

56. Regarding Hesiod's treatment of the F'uries, see Op. 802-4; Th. 182-85, 472-
57. For the euphemistic name "Eumenides" employed on account of mortals' fear, see also
Eur. Or. 37-38. On the general topic, one may consult, in addition, the observation of Burkert
(1985,181) that the ordinary individual's fear of the daimon is indicated by "euphemistic talk about
the 'other daimon' instead of the evil daimon." Find. Pyth. 3.34 (8ai^io)v etepo^), mentioned by
Burkert in this connection, offers a nice illustration of this point. For aW,r) in the phrase aU.r\
(iotpa ("another fate," Soph. Aj. 516) as euphemistic, see Young 1966. Regarding euphemistic
terminology, cf. Nagy's treatment (1974, 260) of Clytemnestra's name, according to which "the
name KX.mai-|ir|OTpa . . . connotes that she is renowned for what she devised, with -ur|CTpa de-
rived from the verb |rr|5o|J.ai. Given the sinister connotations of this verb, the by-form
|a,VT|crcpa may even be the result of tabu-deformation."
36 THE ANCIENT QUARREL

trepidation about the naming of these entities is underscored by Euripides in an

exchange between Orestes and Menelaus following the latter's request for a de-
scription of the phantoms responsible for Orestes' madness (Or. 408-10):

ORESTES: I seemed to see three women, black as night—

MENELAUS: Say no more. I know the spirits you mean. I refuse to speak their name

ORESTES: You are wise. They are awful. 58

As previously discussed, it is precisely this fear that leads mortals to give the dei-
ties an onoma with positive semantic constitution, which encapsulates their hope
that the name's referents will alter their behavior to correspond to the descrip-
tion in question.
Another pertinent example involves Penelope's reluctance to articulate the
onoma of that city the journey to which, she thinks, led ultimately to Odysseus'
death. This deep hesitation leads her to refer on several occasions to "evil Ilium,
that is not to be named" (KdKOiAiov OIJK 6voio,a(JT,f|V, Horn. Od. 19.260,19.597,
23.19).59 According to Brown (1966,199), "the name of Ilios is taboo and not to
be mentioned; so it is evaded . . . by a modification (in this case, compounding)
that leaves it recognizable, but avoids the ill-omened plain word."
In a related type of case, Herodotus declines to mention Osiris' name in con-
nection with a certain activity performed in his honor (2.61.1), saying simply that
"I have already mentioned the festival of Isis at Busiris: it is here that everybody—
tens of thousands of men and women—when the sacrifice is over, beat their
breasts: in whose honor, however, I do not feel it is proper for me to say (tov 5e
TOTCTOVTCU, oi) jiot ooiov e<m A,eyew)." He observes subsequently that when a
corpse is brought to them, Egyptian embalmers display wooden models, vary-
ing in quality, of which "the best and most expensive kind is said to represent a
being whose name I shrink from mentioning in this connection (eivai T.O\> o\)K
ooiov jtme\)(ioa to owouxx ETU TOiomcp 7ipf|yncm ovoua^etv)" (2.86.2). Else-
where, Herodotus refers to that annual festival "on which the Egyptians beat
themselves in honor of that deity whom I must not name in this connection"
(2.132.2) and notes that "in Athena's precinct at Sais . . . is the tomb of one whose
name I prefer not to mention in such a connection. . . . There is a stone-
bordered lake nearby, circular in shape and about the size, I should say, of the
lake called the Wheel on the island of Delos. It is on this lake that the Egyptians
act by night in what they call their Mysteries the Passion of that being whose
name I will not speak" (2.170.1-171.i).60

58. Cf. Orestes' reference elsewhere to these entities as cd dvcovunoi 0eat (Eur. IT 944).
59. My translation.
60. For Herodotus' refusal to discuss the reason behind certain Egyptian religious practices,
see 2.46.2, 47.2. The fact that Osiris is mentioned by name elsewhere in what Selincourt (1972,
 LITERARY ETYMOLOGY 37

One may note, furthermore, the suppression of Odysseus' name in the Odys-
sey by those who wish to protect him. His onoma is connected repeatedly there
with the verb odussomai.61 As Austin (1972, 2) observes, "Homer insists on the
pun, and our only doubt is whether to translate the name in the passive or active
sense, Odysseus the man who suffers pain or inflicts pain, the Hated or the Hater."
On the interpretation of Dimock (1956), Homer has both senses in mind since
"odysseusing" and being "odysseused" are both integral to the construction of
the hero's identity. On the basis of this interpretation, Dimock (1956, 57) pr
poses rendering his name as "Trouble," which may be viewed as having the vir-
tue of ambiguity on account of its easy accommodation to both the passive and
active senses of odussomai.62
Austin (1972) emphasizes the intimate connection between name and per-
son in Homer, noting that "the name is the man in the Odyssey" (3). Brown (1966)
construes Polyphemus' curse on Odysseus as "an instance of the power of the
name" (195): by disclosing his onoma, Odysseus "makes it possible for Polyphemus
to lay a curse on him" (196).63 While it might often make sense for Odysseus
himself to withhold his name out of a concern for his own security, what is note-
worthy in the Odyssey, according to Austin (1972, 5), is that "Odysseus is blessed
with friends who respect the name tabus as much as he. Those closest to him
treat his name as a treasure which must be shielded from vulgar display, protect-
ing the man by repressing the name." Of particular importance in this regard is
14.145-47, where Eumaeus says that he dreads to name Odysseus (TOV . . . eycov
. . . 6vo|a,ri£eiv ai5eo|j.ai) and calls him instead etheion. On the interpretation
of Austin (1972,11-12), "Eumaios will not endanger the life of his master by call-
ing him Odysseus, i.e., the sort of man who outwits or offends the highest de
ties on Olympos, but will name him the man of character (ethos), simply 'the

153 n. i) describes as his "public" capacity, as at 2.42.2 and 2.144.2, suggests that it is the context
that determines whether it is appropriate to provide the god's onoma; in this regard, it is worth
noting that Herodotus typically uses the phrase "in this connection" (em. loiOTJtcpTtpri'yiJ.a'ci) when
indicating that Osiris' name must be omitted (see 2.86.2, 132.2, 170.1).
61. For passages in the Odyssey that underscore the etymological connection, see 1.62, 5.339-
40, 5.423, 19.275-76, 407-9.
62. This positive assessment is shared by Austin (1972, 3), who cites Dimock's rendering of
the name approvingly.
63. As Brown (1966, 201) observes, since the name represents the person, it is crucial that
Polyphemus repeat verbatim Odysseus' identification of himself: "He is pronouncing a formal
curse which can be successful only if properly directed, and consequently he repeats the name
and address exactly as Odysseus has given them." Regarding Odysseus' suppression and eventual
provision of his name to the Phaeacians, see Webber 1989. On Webber's interpretation (13), "it is
always a risk to tell one's name, but, where Odysseus once succumbed to temptation, he will now
act with assurance, usurping the role of bard and taking control of his own kleos." Cf. Austin (1972,
5), who notes that "Odysseus had learned well his lesson; he reveals his name now only to assured
friends."
38 THE ANCIENT QUARREL

good man.' Eumaios's conversation with the stranger is apotropaic first in avoid-
ing the imperiled name, but then, when the name must be spoken, in deprecat-
ing any implications which the name conveys."64
As we saw in other examples, individuals may refrain from using certain
onomata due to fear or reverence associated with their bearers. These emotions
may be present even when terms' descriptive content is not at issue, as in the
case of Osiris. With regard to Odysseus, the fear in question is in the first instance
not for oneself but rather for the name's own referent. This apprehension on
Odysseus' behalf is generated in no small part by what was believed to be the
negative descriptive content of his onoma. While individuals, including Odysseus
himself, may try to avoid any negative effects associated with the name by
refraining from its use, Eumaeus' remarks at Odyssey 14.145-47 are interesting
insofar as they appear to represent a move in the direction of what transpires in
the case of the Eumenides, where a name with negative associations is replaced
by a referring expression with a positive meaning in the hope that reality itself
may be impacted thereby.
Regarding the matter of close ties between elements of language and of real-
ity, for current purposes one may note, finally, the belief in the power of lan-
guage generally and names specifically to impact reality that is manifested in
the use of curses (katadesmoi, or defixiones). Defixiones first appear in Attica,
Sicily, and Olbia in the fifth century B.C. and are dispersed throughout the Greco-
Roman world by the second century A.D.65 Their aim, in brief, is "to influence,
by supernatural means, the actions or welfare of persons . . . against their will"
(Jordan 1985,151). Defixiones from the classical period consist typically of small
sheets of lead, which are inscribed, folded, pierced with a bronze or iron nail,
then deposited in chthonic sanctuaries or buried with the corpse of one who
has come to an untimely end.66 As Faraone (1991, 8) observes, a distinctive fea-
ture of Greek curses is that their authors' aim was to inhibit the performance of
their targets rather than to destroy the individuals themselves.67

64. On the disclosure and concealment of Odysseus' name, see also Higbie 1995,160-76,190.
65. Faraone 1991, 3. Plato evinces a clear awareness of the existence of katadesmoi (see Rep.
364)5-0; Leg. 9333-0); on the importance of the former passage, see Jordan 1988, 276-77. Accord-
ing to Jordan (1980, 227), defixiones remained in use until at least the sixth century A.D. (cf. Jor-
dan 1988, 274).
66. This description is drawn from Faraone 1991, 3. As he observes (1991, 22 n. 5), sometimes
one tablet was pierced with multiple nails; for examples, see Jordan 1985 (hereafter cited as SGD),
40, 41, 72 (numbers here and in subsequent references are to tablets in Jordan's list).
67. According to Faraone (1991, 26 n. 38), the aim of death or destruction is mentioned only
quite rarely in early Greek curse tablets. Defixiones contain neither the names of those initiating
the curses nor defenses of their actions; on this issue, see Versnel (1991, 62-63), who considers
why they are anonymous (cf. his labeling of the provision of authors' names and justifications for
their actions as "nontraditional" features of certain tablets [1991, 68]). On defixiones as not merely
"a lower-class phenomenon," one may consult Faraone 1989, 156 (with n. 20); see also Faraone
1985, 153 (with n. 20) and Jordan 1988.
 LITERARY ETYMOLOGY 39

Faraone (1991) stresses the agonistic social context in which defixiones were
produced. Since agonistic relationships were at issue for the Greeks in many
arenas, it is not surprising that surviving evidence indicates the employment of
curses by individuals with a range of particular aims. According to Faraone (1991,
10-11), with respect to their focus one may divide formulae into four basic types:
commercial curses; those directed against athletes and other public perform-
ers, among them participants in theatrical contests; amatory curses, which
Faraone subdivides into "separation" and "aphrodisiac" varieties; and curses
centering on the judicial arena.68 While some defixiones invoke one or more
divinities associated with the Underworld, whose action is hoped for or antici-
pated, in other cases the individual preparing the curse relies on his own activity
to make it efficacious. 69
As observed, defixiones were used in competitive contexts, where the goal of
the defigens was to impair the object's performance in the pertinent arena. As com-
mentators have noted, extant tablets often target individuals' cognitive and linguistic
capacities, bodily parts, technical expertise, and possessions.70 The more specific
objects of defixiones include the following, quite often in various combinations:71
glotta,72 cheires,7^ podes,74 soma,75 psuche76 nous,77 thumos,7& mneme,79 praxeis,80

68. On judicial curses as composed before, rather than after, the conclusion of trials, see
Faraone rggi, r5, 29 n. 67. For discussion of the differences between this type ofdefixio and judi-
cial prayers, see Versnel rggr. According to Faraone (1991,16), with few exceptions the large group
of published Greek judicial curses date from the classical and Hellenistic periods, and it is likely
that all published Attic judicial curses belong to the former. On judicial curses as an early devel-
opment, see also Faraone 1985, 153 (with n. 19), 154 (with n. 24); in his view (1985, 154), judicial
curses and the civil courts developed simultaneously.
Commercial defixiones generally come from the classical and Hellenistic periods (Faraone
iggr, 10), while those targeting athletes and other performers are generally not found prior to the
second century A.D. (ro-ir). In contrast to aphrodisiac curses, which are strictly a late develop-
ment, separation curses, though infrequent, are evenly distributed across time (11). Faraone (1991,
16, 30-31 n. 76) observes that well-known politicians and orators were often the targets of earl
Attic defixiones, where the curses' aims are either clearly judicial or left unstated.
69. On this topic, see Faraone rggr, 10.
70. On this subject, one may consult, e.g., Faraone rg8g, iggr.
71. The following list of objects and examples is not intended to be exhaustive but rather to
convey a sense of what is at issue in the relevant class of tablets.
72. Wiinsch 1897 (hereafter DTA), 49-50, 52-54, 56, 6r, 66, 68, 75, 79, 82, 84, 87-90, 94~g8
107; SGD r, 46, 95, gg-roo, ro8.
73. DTA 52, 54, 60, 68, 86-87, 89-9°; 93> 96-97-
74. DTA 60, 68, 80, 86-87, 89-90, g3, 96-97.
75. DTA 74, 93.
76. DTA 49-51, 56, 66, 74, 77, 79, 84, 86-87, 89, 93, 96-97, 107; SGD i, 46, loj.
77. DTA 51, 59, 87, 89,107. Cf. the wish expressed in DTA 65 that the curse's targets become
aphrones.
78. DTA 51-53.
79. DTA 61.
80. DTA 56, 63, roo.
40 THE ANCIENT QUARREL

erga,81 ergasia,&2 ergasterion,^ emporion,^ tec/me,85 organa,^ phone?1 /ogos,88

epe,w and rhemata.90
Underscoring the connection between language and reality, Jevons (1908,115)
remarks that the defigens says what is inscribed "and drives a nail or nails through
the leaden tablet bearing the words." Emphasizing the central role played by a
conviction that language shapes reality, Strubbe (1991,41) observes that "the force
of a curse is based on a more general belief in the efficacy of the word," adding
that this effect is enhanced "if the word is spoken by a person of higher status,"
for example, the dead, as was thought to occur when defixiones were placed in
the graves of those who had died before their time (aoroi). Regarding the per-
ceived capacity of language to impact reality, it is worth noting, in addition,
evidence cited by Strubbe (1991,42) that a curse could be viewed as having such
force that it actually became a bad daimon, namely, Am ("Curse").91
While over time the texts inscribed on defixiones become increasingly com-
plex—falling into the basic types previously delineated —most of the earliest Attic
and Sicilian tablets consist exclusively of lists of onomata.92 As Faraone observes
(1985,151), "by fixing the names of the victims with a nail, the defigens attempts
to transfix his or her intended victims."93 The inscription on the tablet of the

81. DTA 53, 56, 63-64, 69, 75, 77, 84, 95, 98, 104; SGD 15, 40, 46.
82. DTA 52, 68-71, 74-75, 86-87.
83. DTA 68, 71, 74-75, 84; SGD 124.
84. DTA 75.
85. DTA 73-74, 87.
86. DTA 73.
87. DTA 61.
88. DTA 94.
89. DTA 56, 84, 95, 98; SGD 40.
90. DTA 68, 93. Regarding the targets of defixiones, see also SGD 3, n, 44, 51, 57, 69.
91. On the connection between speech and action, it is worth singling out one type of bind-
ing formula discussed by Faraone (1991), which he labels "similia similibus" (5). As explained by
Tambiah (1973) in his account of "persuasive analogy," it centers on two objects viewed as hav-
ing both similarities and differences and an attempt to transfer a feature of one object to another
that lacks it (222-23). According to Tambiah (1973, 222), both a linguistic and a mental or physi-
cal act are required for "the full realization of the performance." Strubbe (1991)—whose focus is
imprecations, mainly from Asia Minor, that are intended to protect graves —observes that "the
force of the cursing word could be increased by a variety of rhetorical devices, such as repetition,
rhythm, and the use of triplets. These phenomena are common to both 'magical' and 'religious'
liturgy" (41). An investigation of surviving defixiones reveals a strong concern with rhetorical de-
vices, including repetition and the transposition of letters in adjacent words.
92. Faraone 1991, 5; cf. Versnel iggr, 61. For numerous examples, see DTA 1-39; see also the
list of Jordan (SGD), which includes tablets consisting of names. Faraone observes (1991, ro) that
over three-fourths of published Greek defixiones either consist entirely of names or are so brief
that they provide no clue to their specific aims. In his view (1991, 5), the gradual increase in lit-
eracy in the classical period helps to explain the fact that over time more of the spoken portion of
the curse was inscribed on the tablet itself.
93. According to Faraone (1991, 4), it is not entirely clear whether individuals performed the
binding rituals themselves or hired professionals to do so on their behalf; if the latter is the case,
 LITERARY ETYMOLOGY 4!

target's name was conjoined with the utterance of a verb of binding.94 These
would seem to be individually necessary and jointly sufficient conditions for the
curse's having a chance at success. Clearly, the name's inscription and piercing
are crucial insofar as it is the onoma thus manipulated that establishes the cor-
rect target for the accompanying speech and ritual activity. Since in this context
the name represents the person whose action or performance is to be impeded,
the practice testifies to a strong belief in the existence of a substantial connec-
tion between onomata and their referents. One noteworthy tablet (DTA 57) wa
inscribed with the words:
("I bind the name and tongue, that is, the individual himself'). 95 Jevons (1908,
108) rightly observes that this remark provides strong confirmation of the deep
tie that was believed to exist in this context between an onoma and its bearer. As
he notes, moreover (1908, 107), it is not surprising that the name was used for
the same purpose as a figurine representing the bearer since "the name is, if
anything, even more intimately identified with the man than any likeness of him
can be."96

this would not be due to the necessity of specialized knowledge, since "the act of flattening out
a soft piece of lead and then scratching a name into it certainly did not require much more effort
or technical skill than inscribing a potsherd for a vote of ostracism."
94. On this point, see Faraone rggr, 4-5 (cf. rgSj, rj3 [with n. 21]).
95. I follow Jevons (1908,108) in taking the as epexegetical, although he omits the tablet's
mention of the victim's tongue. For the relevant use of the term onoma, see also SGD 23, 146.
For other pertinent examples, one may consult Jevons 1908, 05.
96. In a variant of the type of binding that centers on the target's onoma, lead "voodoo dolls"
were placed in boxes, with victims' names inscribed on the lids and sometimes on the dolls them-
selves (Faraone rgSj, ijr); on these figurines, see also Jordan 1988. According to Jordan (1988,
275), they were rare in classical Athens. While this form of binding carries the process one step
further, insofar as it involves the construction of figures that were viewed as representations of
the curses' objects, at its core it is a functional equivalent of the approach that centers on names.
 2
ETYMOLOGY AND
THE CRATYLUS' SOURCES
An Investigation of the Literary Tradition's Importance

i. Introduction
Any assessment of Plato's systematic philosophy must devote significant atten-
tion to his philosophy of language.1 The dialogue in which Plato concentrates
most intensively on linguistic issues is the Cratylus. In fact, Plato's Cratylus has
interested those working in the areas of Greek philosophy and philosophy of
language precisely because it is the first text in the Western philosophical tradi-
tion that offers a sustained treatment of such issues.2
The Cratylus poses certain problems regarding the basis on which onomata
("words" or "names") may be judged "appropriate" or "correct." Plato focuses
largely on a thesis according to which names are assigned correctly if their se-
mantic constitution, articulated by etymology, reveals their referents' natures.
Since this is where his own emphasis lies, one cannot develop a defensible inter-
pretation of the dialogue as a whole without taking the etymological section
seriously into account. Despite its dominant presence, some have largely dis-
missed this portion of the Cratylus} Even those who recognize the section's
importance have not fully addressed the issue of why Plato makes an etymologi-
cal approach to correctness the centerpiece of the dialogue.4

1. Strictly speaking, Plato has conceptions of various aspects of language but not what one
would call today a complete philosophy. In using the phrase, I have in mind those conceptions
he does offer, viewed as a totality.
2. On the dialogue's historical primacy, see Kretzmann 1967, 360; Pfeiffer 1968, 59.
3. For interpretations thus inclined, see Shorey 1933; Taylor 1960; Jowett 1892, vol. i; Ryle
1966; Kirk i95r; Mackenzie 1986; Leky [1919] 1967.
4. Those who have emphasized its importance include Gaiser 1974, 25; Baxter 1992,187; Ross
1955, 191; Luce 1964, r4g; Benardete ig8r, 127; Dalimier 1998, 17.

42
 ETYMOLOGY AND THE CRATYLUS SOURCES 43

While long-standing controversy surrounds the interpretation of many Pla-

tonic dialogues, debate regarding the Cratylus has been especially marked by
the range of, and divergence among, competing orientations.5 With regard to
historical sources, interpreters generally emphasize Plato's relation to various
sophists and philosophers. Although reference to these sources often helps to
situate Platonic concerns, explanations of the dialogue's etymological section
that concentrate on these figures encounter numerous difficulties stemming from
the paucity of extant evidence and the fact that such appeals are not adequately
supported by the evidence that has survived.6
Given the problems associated with this search for Platonic antecedents, one
might follow the lead of those commentators who largely dismiss questions of
historical sources in order to concentrate on the dialogue's philosophical con-
tent and import. This focus is endorsed explicitly by Gaiser (1974, n), who main-
tains that "the question about historical preconditions is fundamentally unpro-
ductive" and that "what [Plato] has in mind philosophically in a particular work
emerges not from the historical background . . . but rather from the thinking
that takes place in the dialogue itself." Similarly, Derbolav (1972), calling atten-
tion to the extreme difficulty of identifying the dialogue's sources (27, 31), stresses
that "in fact the historical connections involving the Cratylus are . . . explored
in detail nowhere in the book. . . . The author sees in this no deficiency, but
rather an advantage from the standpoint of the investigation's concentration on
the level of substance" (25). While the general attractiveness of making a virtue
of "necessity" may seem obvious, in the case of the Cratylus historical and philo-
sophical questions cannot be so neatly distinguished; rather, in order to estimate
successfully the dialogue's philosophical content and innovation, one must
achieve the fullest possible awareness of other similar inquiries. 7
Although commentators have attended to a range of possible sources, most
often they either do not mention Greek literature or make only isolated refer-
ences to it. I submit that once one looks beyond those regularly identified as
central opponents of the dialogue's etymological section and attends to literary

5. My treatment of the Cratylus in chapters 2-3 focuses on those points of contention that
are directly relevant to the claims being made in this book. A comprehensive treatment of issues
raised by the dialogue would be well beyond the scope of the present study.
6. The proviso about a scarcity of evidence applies to such individuals as Antisthenes, Cratylus,
Heraclides Ponticus, and Euthyphro. In other cases, prominent among them Prodicus, Protagoras,
and Heraclitus, one has somewhat more evidence to go on. Here, however, commentators as-
sume closer ties to Plato's inquiry than are justifiable. Since Protagoras and Prodicus are those
alleged sources most directly relevant in the immediate context of the dialogue, as Plato frames
his inquiry into orthotes, I concentrate on them in section 2. For comments on other proposed
sources, including the Derveni papyrus, see section 3.
7. This statement does not represent a generalization about Plato's writings in their entirety; it
does not apply, for example, to his handling of the appropriateness ofonomata in the Sophist and
Politicus. Regarding the pertinence of the source issue, the dialogues must be assessed individually.
44 THE ANCIENT QUARREL

practice, the provision of a lack of evidence as grounds for turning away from
the source issue becomes decidedly less compelling. In fact, an investigation of
the matter from this vantage point lends support to the view that the Cratylus,
rather than functioning purely as a work in an insular philosophical tradition
(with sophistic influences), represents a major locus of intersection between the
ancient philosophical and literary traditions. In construing the dialogue as a main
point of intersection between these traditions, I am not claiming that it repre-
sents an amalgam of the two, but instead that it offers fundamental criticisms of
both.8 Scholars have long recognized that when treating ethical questions in the
Republic, specifically the topic of human character in Books 2-3 and 10, Plato
views the poets as direct and central opponents. My claim here is that in what is
ostensibly a quite different arena, namely, his philosophy of language, Plato
adopts a strongly critical stance toward that tradition and views it as a key adver-
sary.9 Most specifically, in contrast to previous commentators, I argue that the
etymological section of the Cratylus targets the literary tradition from Homer
through Euripides.10 My point, it should be emphasized, is not that the literary
tradition is the only relevant backdrop for Plato's discussion but rather that ex-
tant source material supports a view of it as the section's most direct and promi-
nent adversary with respect to etymological praxis.11 It is undoubtedly the case

8. In my view, Plato's critique of the philosophical tradition here is primarily ontological.

The present inquiry focuses largely on the literary tradition's pertinence since that tradition has
received comparatively little attention from interpreters of the Cratylus.
9. As I will argue in chapter 5, there is far more common thematic and conceptual ground
than one might anticipate between Plato's critiques of poetry in the Cratylus and Republic.
ro. In discussing the Cratylus, when using the phrase "literary tradition" I have in mind
Homer's Iliad and Odyssey, Hesiod's Theogony and Works and Days; Pindar's epinician odes; the
extant plays of Aeschylus, Sophocles, and Euripides; and Herodotus. For treatments of literary
etymology from other points of view, see Wilson 1968; Looy 1975. For a discussion of keklesthai
and kaleisthai as used in Greek tragedy and by Plato, see Ruijgh 1976.
11. I thus cannot agree with the claim of Baxter (1992, 107, 127) that much of the time Plato's
targets in the etymological section are '"second-leaguers."' For comments that do not include
mention of the literary tradition, or make only isolated references to it, see, e.g., Fowler 1926, 4;
Taylor 1960, 77, 83 (with n. i); Guthrie 1978,16 (with n. 2), and 1971, 207 (with n. 2), 218; Kahn 1973,
155-56 (with nn. 5-6), 160 n. 12, and 1986, 98-99; Schmitz 1991, 47; Steinthal [1890] 1961, 81, 8
84. Luce (1969,225) refers to Aesch. PV 85-86 and the epitaph of Eutychides and remarks in passing
that "in the detailed etymologies of the Cratylus Plato operates with this traditional Greek con-
ception of proper names, and . . . extends it also to cover general names." The phrase "operates
with" is vague, and Luce refers to a "traditional Greek conception" and "traditional Greek view"
(225) rather than to the literary tradition specifically. Pfeiffer (1968, 61) mentions but does not
discuss the "task of , familiar from Homeric times"; it thus remains unclear how Plato's
inquiry is supposed to be related to these earlier reflections. Gaiser (1974, 51) recognizes that poets
engaged in etymology but does not pursue the matter. Dcrbolav (1972) refers to a "chorus of poets
and philosophers" who employed this procedure (34) and says that Plato stands in the traditio
"of an etymological practice that was at least occasionally utilized by poets and philosophers"
(69 n. 2); however, Derbolav neither identifies the members of this group nor specifies the perti-
nence of their remarks to the dialogue's interpretation. Diels (1910, 3-7) sees the relevance of
 ETYMOLOGY AND THE CRATYLUS SOURCES 45

that having further evidence from various authors would allow one to formulate
a more complete account of the Cratylus' sources in this regard. One must not,
however, be led by the existence of gaps in what is available to neglect the com-
prehensive study of all pertinent material that has survived.

2. Orthotes Onomaton: The Cratylus Engagement

with Literary Practice
Back in 1867, Grote judged unacceptable the "modern discovery or hypothesis"
that the sophists are the targets of Plato's investigation of the correctness of names,
insisting on the absence of proof that either Prodicus or Protagoras was concerned
with etymology (521).12 Yet numerous scholars continue to espouse the view that
Grote attacked. Kerferd (1981, 77), for example, asserts that the stance on orthotes
developed in the Cratylus represents "Plato's contribution to the problem which
he inherited from the sophists," and Guthrie (1978, 4) believes it "a reasonable
conclusion that Plato found Cratylus the Heraclitean a suitable character through
which to criticize prevailing beliefs of the Sophists about the relationship be-
tween words and reality." Weingartner (1973), in turn, claims that Plato chal-
lenges the views of such "professional predecessors and contemporaries" as
Protagoras (18) and asserts that "in all probability Plato uses the opportunity
provided by the etymology section to poke fun at various other theorists of lan-
guage of his own time and of the recent past. Perhaps the great Prodicus' fifty-
drachma course is satirized more than we shall ever know" (40).

poets, among others, to discussions of Greek etymology but does not single them out in his com-
ments on the Cratylus, which focus instead on Heraclitus and later Heracliteans. Reeve (rggS,
xxxi) mentions examples of etymologizing in Greek tragedy but does not build on these observa-
tions in subsequent remarks on the dialogue's target, claiming instead (xxxii) that Plato employs
etymologies "to criticize or parody specific individual thinkers who use etymological arguments
in philosophy without showing any awareness of the problematic presuppositions of doing so."
Meridicr (1931) cites literary passages but believes (30) that the target of the dialogue's polemic is
certain contemporary theories of language. Warburg (1929) evinces some awareness of certain
poets' use of etymology, yet contends that the Cratylus targets Heraclides Ponticus. It is not evi-
dent, moreover, precisely what role Warburg views Greek literature specifically as having vis-a-
vis the Cratylus; clarity on this point would require a systematic treatment of the literary tradition's
techniques and assumptions from Homer onward. Although Baxter (1992) remarks on the liter-
ary tradition, he identifies it as merely one among numerous other likely or possible opponents
of the etymological section. In addition to the fact that he treats only Homer and Aeschylus, Baxter
tentatively associates certain of the Cratylus' etymologies with other sources although they can
be linked clearly to poets, as when the Hera-aer derivation is tied to Empedocles (124, 125), and
the Demeter-didousa has meter analysis is linked to the Derveni papyrus specifically or Orphic
'"scholars'" more generally (132).
12. For an emphasis on the lack of evidence to support the claim that Prodicus and Protagoras
pursued etymology, see also jvleridier rgjr, 42-44.
46 THE ANCIENT QUARREL

In setting the stage for his treatment of etymology in the Cratylus, Plato refers
to certain conventional authorities, most notably Prodicus and Protagoras, to
whom orthotes onomaton was indeed a concern. Plato's mentions of the two
sophists, combined with their known interest in this topic, may appear to consti-
tute strong support for the claim that there is a central sophistic influence oper-
ating in the dialogue. This view of the sophists seems plausible, however, only
as long as one does not distinguish clearly between different senses of the term
orthotes as it appears in the phrase orthotes onomaton.n Once one sees how
Prodicus and Protagoras treated orthotes, and that Plato's interest assumes a
quite different form, it becomes evident why the two sophists are of only lim-
ited pertinence to the Cratylus. My contention is thus not that they are irrele-
vant thereto but rather that their importance vis-a-vis the dialogue has not been
properly located. Plato and the sophists share a common literary heritage. Plato
is dissatisfied with the way in which they utilize this material in their treat-
ments of linguistic issues and undertakes his own exploration of the literary
tradition. In so doing, he does not take sophistic treatments of correctness as
his point of departure. Rather, as I will argue, Plato ties his inquiry directly to
literary practice.
At the outset of the dialogue, Socrates mentions Prodicus, commenting on
his own failure to attend the latter's thoroughgoing fifty-drachma epideixis "on
the correctness of words" (nepi OVOIHTOOV opeonyrcx;) (jS/fbz-ci). The transla-
tion of Jowett (1892, vol. i, "If I had not been poor . . .") actually gives poverty as
the reason for Socrates' lack of attendance although it is not present in the Greek,
which simply consists of a counterfactual statement with the protasis "If I had
heard Prodicus' fifty-drachma epideixis"
14
Jowett's rendering thus fosters
the misleading impression that the content of such an epideixis would be of
direct relevance in the present context. Although Taylor (1960, 77 n. 2) rightly
points out that Socrates' poverty is not the reason given for his lack of familiar-
ity with this specialized treatment, he does not suggest what the actual ground
might be.

13. One may underscore the same point in other terms by noting that since the ontological
relation is one of genus to species, the strongest defenses of relationships will naturally occur at
the species level. Furthermore, the mere invocation of a clash between "nature" and "conven-
tion" does not require that the content of the opposed views be sophistic in inspiration. Guthrie
(1971, 55) describes nomos and phusis as "catch-words" of Greek reflection in the fifth and fourth
centuries. For the suggestion that this opposition grounds the Cratylus' discussion, see Guthrie
1971, 208; Mackenzie 1986, 127; Pfeiffer 1968, 63; Taylor 1960, 77. On this issue, I concur with
Robinson (1969, m-r6), who argues that one is not confronted here with simply another instance
of the familiar nomos-phusis controversy and that the divergence involved is more than merely
terminological.
14. Aristotle too refers to the fifty-drachma version (Rhet. 14151)15—17). As Guthrie (1971, 42
[with n. i|) observes, there is some controversy over whether a single lecture or a series thereof
was involved.
 ETYMOLOGY AND THE CRATYLUS SOURCES 47

On Plato's own account in other dialogues, Prodicus' interest in the issue of

correctness lay in making fine distinctions between terms with different yet closely
related senses and occasionally in differentiating senses of a single term. In the
Protagoras, for example, Prodicus is said to distinguish between the terms
"impartial" and "undecided"; "debate" and "dispute"; "esteem" and "praise";
"enjoyment" and "pleasure"; and "will" and "desire" (3373-0, 34oa-b).15 In his
view, the comprehension of words' meanings reflected in prevailing usage was
often not sufficiently subtle. A superior understanding was possible, however,
and would issue in corrections of deficient usage; this idea of the revision of ordi-
nary practice emerges clearly in remarks attributed to him on the proper employ-
ment of the term deinon (Prt. 34ia-b). Prodicus thus raised questions of mean-
ing without reference to words' constitution. While his concern with words
certainly fell under the rubric of "correctness," Prodicus' own treatment of
orthotes is strikingly different from Plato's, and this fact sheds light on Socrates'
claim (Cra. 384^ not to have heard Prodicus' detailed epideixis on the subject;
Plato's point, in my view, is that a consideration of Prodicus' own orientation is
not relevant to the current enterprise. That Socrates has listened to the far more
superficial one-drachma treatment (ci) is evidence of a limited, general point of
common ground, insofar as Prodicus too was associated with the topic of cor-
rectness. Notably, however, even at this initial stage Plato indicates through what
is said about Hermogenes' name (in 3840, which is based on 383^ that he con-
ceives of orthotes along the lines of descriptive content or semantic constitution,
hence in a substantially different light from both Prodicus and, as we shall see,
Protagoras.16
Protagoras was concerned with dividing logos into types. In this connection,
he referred to inaccuracies in others' usage due to shortcomings in their under-
standing of the relevant distinctions. This occupation is stressed in Aristotle's Po-

15. One may consult, in addition, Prt. 358a-b, Meno 756, Euthd. 2776-2783, La. igyb-d, Chrm.
r63b-d; Arist. Top. II H2b. In the Euthydemus passage, one finds the comment that "according to
Prodicus, one must learn, first and foremost, about the correctness of words
(27763-4). The content given by this sophist to the notion of correctness is made
clear in the remarks immediately following, which treat different senses of to manthanem.
16. At 383^ Hermogenes conveys to Socrates the view of Cratylns that Hermogenes' name
(which means "son of Hermes") is not properly assigned to him and expresses his own uncer-
tainty regarding the interpretation of this pronouncement (3843). On Socrates' highly compressed
account (3840), which I expand here to fill in the gaps, Crstylus means that Hermogenes is not
fittingly tied to Hermes as the name suggests: Since the god brings luck to human beings, one
would expect that those linked to him via their names would not be lacking in good fortune.
Because Hermogenes falls distinctly short in this department, it is inappropriate for him to bear
a name whose descriptive content suggests otherwise. Literary authors too offered judgments of
names' inappropriateness. A prominent example, noted in chapter i, is Cratus' insistence to
Prometheus that "the gods named you wrongly when they called you Forethought (yex)8(ovuu(0(;
yon yourself need Forethought to extricate yourself from this
contrivance" (Aesch. PV 85-87).
48 THE ANCIENT QUARREL

etics (145613), where Protagoras is alleged to have faulted Homer for opening the
Iliad with a command when the poet thought he was uttering a prayer. 17
Protagoras is also said to have divided nouns into three genders, prescribing
adjustments in gender where fitting (see Aristophanes, Clouds 658-91; Aristotle,
Sophistici Elenchi 1730). In the passage from Sophistici Elenchi, Aristotle reports
Protagoras' insistence that the words for "wrath" (menis) and "helmet" (pelex),
currently feminine in gender, be changed or corrected to the masculine.18 Once
again, these observations and prescriptions do not involve the treatment of words'
constitution.
Protagoras, like Prodicus, is introduced early in the Cratylus (38566), and on
the strength of this and of his reputation for treating "correctness," one might
conclude that he is of central relevance. Yet Protagoras is not mentioned in
connection with an interest in onomata in general or orthotes in particular; rather,
Socrates invokes Protagoras' "man-measure" doctrine in his attempt to make a
stubborn Hermogenes realize the ontological position to which he commits
himself in espousing an extreme view on the matter of appropriateness (38504-
386d2). Protagoras is reintroduced as Socrates —at Hermogenes' request—pre-
pares to undertake a sustained exploration of what the "correctness of words"
consists in (39066-39^9).19 Here Protagoras is clearly associated with the issue
of correctness (39^2-5). At this pivotal juncture, however, his own authority or
relevance, along with that of sophists in general, is decisively rejected (39ic-d).
In contrast to the earlier comment about Prodicus (384b-c), poverty is actually
mentioned: it is referred to as what prevents Hermogenes from consulting the
sophists —specifically Protagoras —directly (39^9^5). Yet Hermogenes pro-
ceeds to indicate that what ultimately and categorically precludes his turning
to Protagoras is not a lack of means but his rejection of that sophist's construc-
tion of reality: following Socrates' "recommendation" that he learn Protagoras'
view secondhand from Callias, his own brother, Hermogenes observes that it
would be an odd thing indeed "if, not accepting at all the Truth of Protagoras, I
were to treat what is said therein as being at all worthwhile" (39107-9). Thus,

17. Having reported this objection, Aristotle judges it to be inappropriate. According to

Protagoras, the study of poetry constitutes the most important dimension of education (Prt. 33866-
33931); at issue is not the wholesale acceptance of poets' claims and assumptions but rather a
capacity for the critical assessment thereof (for an emphasis on the ability to subject their remarks
to evaluation, see 33931-3).
18. Proposed grounds have included both the notion that the referents of these onomata are
strongly associated with the male sex and a concern with morphological consistency (on this point,
see Kerferd 1981, 69).
19. According to Silverman (1992, 44), already in 389-390 Plato operates with the view that a
name's constitution is "irrelevant to its embodiment of the proper form of its nominatum": "The
argument. . . suggests that names . . . are used by a name-giver as proxies for that item which he
picks out via a definition" (43; see also 54). On my interpretation of the passage, Plato's point is
not that it is irrelevant whether there is a match between words' semantic constitution and their
referents' natures but that different sets of linguistic components can achieve that end.
 ETYMOLOGY AND THE CRATYLUS SOURCES 49

poverty is not depicted in the case of either sophist as what prevents direct con-
sultation of his stance on the correctness otonomata.
This rejection of the sophists' pertinence occurs precisely when Socrates, hav-
ing established the apparent tenability of the "natural-correctness" thesis, seeks an
appropriate point of departure for the inquiry into orthotes.20 This fact, combined
with the preceding treatment of Prodicus and Protagoras, suggests that the soph-
ists are not key to the enterprise as Plato conceives it. Where, then, does Socrates
turn in seeking a foundation for what follows? Precisely to the literary tradition of
the eighth through fifth centuries: Hermogenes' rejection prompts Socrates'
response that he "must learn from Homer and the other poets" (39icio-di).
Commentators regularly dismiss this comment by invoking the hostility
expressed elsewhere by Plato toward the literary tradition. Thus Friedlander
(1964,204) says that following the rejection of Protagoras at jgic, Socrates "sug-
gests that we can learn from Homer what he has to say about language; yet we
know since the Protagoras that an exegesis of poetry does not yield knowledge."
In a similar vein, Weingartner (1973, 38) dismisses Socrates' comment as invok-
ing "a fine source of knowledge for the Plato who banned Homer from his
Republic!"21 The fact that Plato elsewhere rejects poets' creations as sources of
knowledge does not in itself justify the assumption that the literary tradition could
not serve as a foundation for his discussion in the Cratylus. That he takes this
tradition quite seriously indeed is evident from the fact that the Republic devotes
so much space to delineating where author and creation go awry. If Plato is
capable of engaging the literary tradition once with this degree of earnestness,
there is no obvious reason why he could not do so in other places. As one reflects
on particular dialogues in this regard, the crucial consideration should be, not
whether he has elsewhere challenged its authority, but whether it and the text in
question supply evidence of a deep connection. While the Cratylus will ulti-
mately reject the idea, prominent in literature, that etymologies ofonomata dis-
close their individual referents' natures, dismissal of the literary tradition's per-
tinence at this juncture is premature: rather than signaling his own embrace of
its approach, Plato is setting the stage for his argument that such a reliance on
etymology and privileging of individual "natures" are fundamentally misguided.
Interpreters have sometimes gone even further, ,,using Cratylus 3gic-d as
grounds for making speculative inferences about the content of sophistic inquir-
ies. Kerferd (1981, 75) construes the passage as offering "sufficient grounds for
us to conclude that in his work On Truth Protagoras had in fact discussed the
Tightness of names, and the natural way to read the passage is to suppose that
Protagoras had himself in some sense and in some degree given expression to a

20. For this as the discussion's focus, see 39^4-5.

21. See also Meridier 1931, 15-16; Rumsey 1987, 392, 402 n. 16. Similarly, without considering
the possibility that the passage has greater import, Jowett (1892, vol. i, 260) remarks that, as in the
Republic, Socrates makes an ironic appeal to Homer.
50 THE ANCIENT QUARREL

belief in the doctrine of natural Tightness." According to Guthrie (1971, 206),

Homer's identification as an authority provides support for the claim that the
investigation of orthoepeia ("correct diction"), like that of the "correctness of
words," probably encompassed reflection on words' natural appropriateness (for
Plato's correlation of orthoepeia with Protagoras, see Phdr. 267^-6). Such a
construction of Plato's reference to Homer is puzzling given the fact that it is
precisely Protagoras' authority that has just been rejected (3910). Moreover, far
from dismissing literary sources, Hermogenes is keenly interesting in learning,
first of all, what Homer—the literary tradition's oldest and most esteemed rep-
resentative—had to say on the matter. The Cratylus offers a remarkable num-
ber of etymologies, and Baxter (1992, 163) is surely right to stress the degree to
which the analyses themselves "remain Plato's creation." The key challenge
confronting interpreters is to determine who provides the most crucial impetus
to Plato's reflections. With regard to the making of this assessment, it is of essen-
tial import that his own exploration ot orthotes begins with an invocation of the
literary tradition.
The initial objects of inquiry are Homeric instances of different onomata used
by gods and humans with reference to the same entities (39^4-392^). The as-
sumption is that those appellations employed by the gods are correct. There are
several passages in which Homer says that men give an entity one name, while
the gods employ another (II. 1.403-4, 2.811-14,14.290-91, 20.73-74); Plato men-
tions all of these instances except the first. The issue of names' having a divine
source first arises here, as Plato makes the transition to poets in grounding his
approach to orthotes.22 To claim that particular assignments are superior just
because they originate with divinities leaves no room for a sustained investiga-
tion of the sort that Plato wishes to pursue. While these instances serve to intro-
duce Homer—and by extension the literary tradition more generally—as hav-
ing treated correctness, it is the ensuing discussion of the names of Hector and
his son that typifies the species of orthotes of central concern to the Cratylus.2^
In fact, Plato uses these onomata, which were salient objects of Homeric
analysis, to underscore what is meant by orthotes in the present context (392b-

22. In his subsequent discussion of names, Plato labels appeals to divine sources as one key
excuse for an inability to offer accounts of onomata that are genuinely explanatory (4263). The
appeal at jgrd-jgzb is somewhat different, insofar as it involves two perspectives, mortal and divine,
the latter of which is identified as superior. This dual perspective, albeit reconstituted, will re-
emerge at the close of the dialogue, where Plato contrasts a philosophical orientation toward fit-
ness having Forms as its reference points with that inferior one, endorsed by ordinary mortals,
which is criticized in what precedes.
23. According to Williams (1982,84), although the first cases treated in the dialogue "are proper
names . . . this is not the basic case, and the theory applies to general terms; indeed, it applies to
proper names because it applies to general terms." Williams does not specify the perspective from
which general terms qualify as the "basic case." For the tradition that grounds Plato's inquiry,
proper names are basic. As Kahn (1973, 159) observes, while the onomata treated are not limited
to proper names, "they remain throughout the typical or paradigm case."
 ETYMOLOGY AND THE CRATYLUS SOURCES 51

393b; cf. 394by-ci). The more insightful individuals, identified by Plato as the
men of Troy, are alleged to have called the boy "Astyanax"; the less insightful,
here Trojan women, are supposed to have employed the name "Scamandrius"
(392cio-d3). Plato notes that the names "Astyanax" and "Hector" are very close
in semantic constitution—"They mean nearly the same thing, both being names
descriptive of a king," 393a6-y24—and connects the latter to the verb echo ("hold,"
"protect," 393bi). Hector's name, like that of his son, is treated as fittingly as-
signed, and Plato concludes that light has been shed on Homer's belief "about
the correctness of words" (393b3~4).
In the Iliad, it is actually Hector who calls his son "Scamandrius," while others
prefer the name "Astyanax," "since Hector alone guarded Troy" (6.402-3; see also
22.506-7 and 24.499-501). Homer's use of the term epiklesis ("additional name,"
22.506) to identify the status of "Astyanax" leaves no doubt that "Scamandrius" is
the child's primary appellation. One can see why Plato would wish to reassign
the name "Scamandrius" to a different source: attributing a preference for
"Scamandrius" to Hector would indicate, as in Homer, that it is primary, and it
would be odd indeed for the child's own parent to have made what Plato iden-
tifies here as the inferior assignment. It is essential that the name "Astyanax" be
seen as superior since only it can be investigated readily by etymology. The only
obvious treatment of "Scamandrius," in contrast, is via eponymy (i.e., the indi-
vidual as named after the river Scamander); while Plato does not cite this epony-
mous derivation explicitly—which he need not do for his audience to be aware
of it—one of the foregoing examples, taken from Homer, mentions the river twice
by name (Cm. 3916-3923). Also key to the preference for "Astyanax" is the fact
that this name accommodates itself to the ties in semantic constitution that Plato
wishes to forge with "Hector." Plato's misidentification is doubtless intentional
and concerns a well-known passage with which his audience would be familiar.
The presence of this "mistake" supports the view that Plato is not merely repeat-
ing his literary sources but drawing on them for his own philosophical purposes.25
Having provided examples of Homer's handling ofonomata and been assured
by Hermogenes that he is on the right track, one by one Socrates subjects the
names of members of the Mycenaean House of Atreus to etymological analysis:
Orestes (3946), Agamemnon (3953-^, Atreus (39513-0), Pelops (395c-d), and

24
25. In his comments on this passage, Fowler (1926, 37) refers only to II. 22.506-7 and does not
note Plato's misidentification of the source as Homer had given it earlier in the poem. Shorey
(1933, 261) mentions 6.402-3 but not the shift in attribution. Meridier (1931,16, 64), in turn, notes
Plato's deviation from Homer but interprets it differently; on this issue, see also Rosenstock K)f)2,
400-402. One finds a similar move in the Timaeus (400-413), where Plato departs from Hesiod's
genealogy. In this case too the relevant literary passage (Th. 116 ff.) would be known to Plato's
audience. I am grateful to the late Wilbur Knorr for drawing my attention to this example from
the Timaeus. See Reeve (1998, xxv-xxvi) for comments, from a different perspective, on Homer's
pertinence to Plato's framing of the etymological inquiry.
52 THE ANCIENT QUARREL

Tantalus (^gjd-e). Finally, he turns to Zeus, the supreme Olympian deity, who
was supposed by tradition to be Tantalus' father (3956-396^. While Agamemnon
and Menelaus play important roles in the Iliad, they do so as Achaean warriors
battling the Trojans and not as members of the House of Atreus per se; it is pre-
cisely among the Greek tragedians that the affairs of the royal house itself play a
central role. These etymologies lend additional momentum and significance to
Socrates' turn to the literary tradition. Moreover, in what follows, Hesiod is given
prominence, Homer is cited repeatedly, and both the practice and the practi-
tioners of tragedy are referred to.26
Plato's choice of onomata subject to etymological analysis further strength-
ens the Cratylus' ties to the literary tradition. Benardete (1981,136) has claimed
that "'Aphrodite' is the only [divine] name whose etymology is traditional." As
will become clear, this is not the only case in which Plato incorporates inher-
ited derivations. In the following illustrations involving divine onomata, I move
from the relevant literary treatments to Plato's constructions.27
Both Homer and Euripides underscore the descriptive content of the names
"Hector" and "Astyanax," along with the semantic relationship between those
onomata. Thus, Andromache's comments to her dead husband include the verb
echo: "I think that our son will never come of age, for before then head to heel
this city will be sacked, for you, its defender, are gone, you who guarded the city,
and the grave wives, and the innocent children
(Horn. II. 24.727-30; on Hector's protective role, reflected in
his name, see also 24.499 ancl Eur. Tro. 590). The Trojans call Hector's son
"Astyanax" ("Lord-of-the-city") on account of Hector's civic function (Horn.
II. 6.402-3, 22.506-7; Eur. Tro. 1168 with 1217). Plato gives a pivotal role to
the names of Hector and his son insofar as he uses them to clarify his own
construction of orthotes. As we saw, Socrates underscores the closeness in
semantic constitution of the names "Astyanax" and "Hector" and follows Homer
in tying the latter to echo. Plato's correction of Homer through his elevation
of "Astyanax" over "Scamandrius" is necessitated by the Cratylus' different
agenda and illustrates Plato's willingness to adapt his literary sources to his own
ends.
As noted in chapter i, Dis and Zeus, the two forms of Zeus's name, were
popular objects of literary reflection: authors trace Dis to the preposition dia (Hes.
Op. 1-4; Aesch. Ag. 1485-87) and connect the form Zeus with the verb zen (Find.

26. For appearances of Hesiod's name, see 39604, 39765, 40286, 40607, 42831. Regarding
Homer, see 391010, d2 (followed by quotations in 39165—6, 39235), 39289, cio, d5, 64, 39332, bj,
40236, b4, 40739, 40834, 41002, 41708; for Homer quoted but not mentioned by name, see 4oyd8-
9 (with in place of 41532, 42804-5, dy-o. Concerning tragedy 3nd tragic
poets, see 4o8c-d, 425d5 (also worth noting in this regard are 41405-6 | and 4i8d4

27. Since I have alresdy discussed literary authors' reflections on these names, my remarks
here will be brief; for fuller discussion, see chapter i.
 ETYMOLOGY AND THE CRATYLUS SOURCES 53

Isthm. 3.4-5; Aesch. Supp. 584-85; Eur. Or. 1635).28 In the Cratylus (3956-396^,
Plato provides a single analysis encompassing both forms and derivations: on
his account, Zeus is the god "through whom all living creatures always have life"
3g6bi). Hence, his treatment of the
god's onoma not only forges common ground with several poets regarding a
particular etymology but also shows that Plato could beat the literary tradition
at its own game —if only he were interested in playing!29 Thus, as in the case
involving Hector and his son, here too Plato relates himself directly to his liter-
ary sources: whereas in the previous instance Plato's innovation consisted in the
revision of Homer's assignments, here it involves a new synthesis of familiar ele-
ments. Grote (1867, 526) claims that no etymology offered in the dialogue "is
more strange than that of and attempts to dissipate this alleged
strangeness by citing its reproduction in De Mundo (401313-15), a pseudo-Ar-
istotelian treatise, and use by the Stoic Zeno (on the authority of Diogenes
Laertius j.i/^j).30 Yet references to later practice cannot diminish the supposed
oddity of Plato's analysis. On the present interpretation, it is precisely recourse
to etymologies previously offered by poets (i.e., Hesiod, Pindar, Aeschylus, and
Euripides) that makes his approach intelligible.
Literary reflections on the name "Demeter" indicate the derivation "Mother-
Earth." Thus, Euripides calls Demeter "the Mountain Mother of all the gods" (Hel.
1301-2; cf. 1320,1340, and 1356), noting that while she grieved over her daughter,
"earth, green gone from her fields, would give food no more to mortals in the sown
lands" (1327-28; see also Eur. Phoen. 683-86, where Demeter is described as "Earth,
nurse of all," and Bacch. 274-76). In the Cratylus, Plato adopts the same deriva-
tion, in a formulation that is particularly reminiscent of Euripides' Helen: "The
goddess Demeter appears to have received her name from the fact that she gives
food to mortals, thus providing for them like a mother" (Ar|ur|Tr|p (lev <|>atveTca
404b8-o).
Homer's treatment of Hades as the "Unseen-one" (II. 5.844—45) underscores
the god's special capacity to avoid detection. In the Cratylus, Plato wishes to revise
this analysis, replacing it with one that will not elicit mortals' fear. He begins by
having Socrates make the following observation: "People seem to assume that
the name 'Hades' is derived from the 'unseen' and due to their fear of
this appellation they call the god 'Pluto' instead
(40335-8). Having mentioned the traditional derivation of
"Hades" from aides, Socrates proceeds to offer a different interpretation: "The
name 'Hades,' Hermogenes, does not come from the 'unseen,' but is instead far
more plausibly traced to the god's knowing all fine things" (404^-3).

28. On the tic between Zeus and zen, cf. Aesch. Ag. 168-75 w''n Quincey 1963, 148.
29. For a recent argument that stresses the agonistic character of the etymological section as
a whole but that does not treat the source issue in detail, see Barney 1998.
30. With regard to De Mundo, I have consulted the Bude edition of Lorimer (1933).
54 THE ANCIENT QUARREL

Aeschylus traces the name "Cronus" to the verb kraino ("accomplish"), as

evidenced by Prometheus' remark that Cronus' curse will be consummated
(kranthesetai) unless Zeus solicits and receives Prometheus' aid (PV 910-11).
Plato's analysis, which is to all appearances more complimentary, treats the name
as signifying koros, "not as in 'child' but rather as in 'the pure and unmixed char-
acter of the mind itself"
Cra. 396b6-7).31
According to Herodotus (2.52.1), the Pelasgians used the Greek word theoi to
refer to the gods because they had put everything in order (kosmoi thentes ta
panta pregmata). Herodotus claims, in addition, that initially the Pelasgians
sacrificed and prayed to the gods without any distinction of appellation and that
the names of individual gods were introduced only later into Greece from Egypt
(2.52.1-2). On Plato's account (Cra. 39yc-d), natural and heavenly bodies were
the original deities and received the collective appellation theoi due to their being
in perpetual motion: "People saw all of them always moving in a circle and, as
it were, running and due to these entities' characteristic nature of
running they called them 'gods'
(39yd2-5). The Olympians, when later introduced, were covered by the same
group onoma (d5~6). While they differ in the specifics of their analyses, both
Herodotus and Plato thus note an early stage of religious belief and practice when
the term theoi was employed but the names of the familiar anthropomorphic
deities were not yet in use.
As we saw in chapter i, in literature a speaker may trace a divinity's name to an
element over which its referent has control. Thus, Homer derives Hera's name
from aer (II. 5.775-76,14.282; see also Eur. Hel. 32-35 and 241-51). Notably, the
derivation from aer is one of two analyses of the goddess's name that Plato offers
in the Cratylus: "Perhaps the legislator, engaging in astronomical speculation, and
being at the same time rather cryptic, linked 'Hera' to 'air.' You might recognize
this if you tacked the beginning of the name closely onto the end, as would hap-
pen if you repeated the name often in rapid succession i

40402-5). While
one might assume the "legislator" engaged in such thinking to be a philosopher
of nature (see the speculation of Baxter [1992, 124, 125] about a connection to
Empedocles), extant evidence thus allows one to identify the figure in question
here as a poet.32

31. I omit here the words which were added in angle brackets following
by the editors of the new OCT.
32. As Guthrie (1965, 144-46) notes, long-standing controversy surrounds the correlation of
the divine names mentioned by Empedocles (DK 6) with the four elements. In the case of Hera,
commentators diverge on the issue of whether the tie envisioned is to earth or to air; Guthrie
himself prefers the latter interpretation (146). Seeking to buttress Guthrie's claim, Sprague (1972)
 ETYMOLOGY AND THE CRATYLUS SOURCES 55

Literary treatments of the name "Apollo" often stress the god's role as the cause
of destructive action by tying his onoma to the verb apollumi ("destroy"). In
Aeschylus' Agamemnon, Cassandra laments her fate as follows: "Apollo, Apollo!
Lord of the ways, my destroyer. You have undone me once again, and utterly"

1080-82; see also Ag. 1085-86 and Eur. Or. 119-21, 954-56)." Hoping to direct
the god's power in a more salutary direction, Jocasta proposes the name's con-
nection to the verb lud ("release," Soph. OT 919-21), and a Euripidean Chorus
asks him to serve as Admetus' luterios ("redeemer") from death (Ale. 224).
Forging a parallel with his earlier derivation of "Hades," Plato prefaces his
own analysis with a reference to the belief, underscored in Agamemnon and
Orestes, that the name "Apollo" has a negative signification: "In many people,
the god's name generates fear, as indicating something dreadful" (Cra. 40461-2).
Having introduced this familiar view, Plato proceeds immediately to correct it.
On his revised account, the name illuminates all of Apollo's positive capaci-
ties: indeed, "there is no single name better suited to express the four powers
of the god, as it captures and illuminates in some way those involv-
ing music, prophecy, medicine, and archery" (40468-40534). Plato's summary
of the results of his own inquiry testifies once again to the depth and persistence
of mortals' fear: "Some still maintain their suspicion that the name signifies
destruction because they do not consider its actual import, which, as I was just
now saying, consists in its capturing all of the capacities of the god: the one who
is single always-shooting ( purifying
and moving-together (40562-40633). Interestingly, Plato's dis-
cussion of Apollo's role as purifier includes a compound of luo (apoluon, 4O5by;
cf. apoluseis, bg), aligning itself thereby with the more positive emphasis of
Oedipus Tyrannus (lusis, "deliverance") and Alcestis (luterios, "redeemer").
Hesiod and Euripides offer radically different interpretations of the name
"Aphrodite": while the former derives it from aphros ("foam," Th. 197-98), the
latter traces it to aphrosune ("folly," Tro. 989-90). In the Cratylus (4o6c-d), Plato
opts explicitly for Hesiod's derivation: "Concerning Aphrodite there is no rea-
son to take issue with Hesiod. One can instead simply agree that the goddess
received her name due to her genesis from the foam"

Homer's explanation of Athena's allegiance to Odysseus (Od. 13.330-32)

focuses on their common intellectual gifts, with the pivotal terms being noema

invokes Cra. 4040. While Plato's derivation of Hera from aer is pertinent to the discussion of
Empedocles' assignments, more persuasive evidence that this is the right correlation is the fact of
its existence prior to Empedocles' own reflections.
33. Mortals' fear of the god's destructive capacity is also reflected in Aeschylus' derivation of
the epithet "Phoebus" from phobos (Pers. 205-6).
56 THE ANCIENT QUARREL

and anchinoos (for stress on her intelligence, see also Hes. Th. 886-96; cf. Th.
924 and Find. Ol. 7-35-37). Notably, Plato ties the name "Athena" to the term
nous (Cm. 4071)2) and begins his analysis (4073-0) with a mention of Homer.
Aeschylus derives the name "Ares" from the noun ara ("curse," Sept. 944-46).
Plato too treats the god's onoma, offering an analysis that moves along different
lines: "Ares may receive his name, if you will, from his strength and courage or
from his hardness and unalterability, which is the meaning ofarratos"

Cra. 4O7di-3).
As noted in chapter i, the literary tradition treats justice as a female deity.
Both Aeschylus (Cho. 948-51; Sept. 662) and Hesiod (Op. 256) trace the name
"Dike" to the status of its bearer as Zeus's daughter (her onoma is a compressed
version of Dfos &ora).34 In etymologizing "justice," Plato refrains from personi-
fication and treats instead the terms dikaiosune and dikaion (Cm. 4i2c~4i3d).
While dikaiosune is easily disposed of as dikaiou sunesis ("comprehension of the
just," 412C7-8), dikaion is far more challenging: although a connection to diaion
("that which pierces") is generally accepted (41263-41331), on a more specific level
constructions of justice diverge. According to Plato, at this point it becomes a matter
of interpretation, and hence of disagreement, precisely how the referent of dikaion
will be construed. Indeed, every thinker offering an account of the cosmos that
privileged one or more "explainers"—for instance, air, fire, to apeiron, mind, love
and strife—would have very definite views on this topic. I submit that Plato uses
the analysis of dikaion, whose referent plays a crucial role in his own ontology, as
a vehicle for suggesting that etymologies of terms' descriptive content cannot re-
solve disputes about phuseis. The existence of such pivotal defects will lead him
to conclude, and state explicitly, at the end of the dialogue (438-440) that what is
required is movement in the opposite direction, namely, from natures to onomata.
The analysis of dikaion thus represents a crucial turning point in the discussion:
by clearly marking the inadequacy of etymology from a philosophical perspective,
Plato takes an important step in the direction of what will have become by the
dialogue's end a bridge to his own theories oforthotes onomaton.
Turning briefly to the names of heroes, one may note that on the analysis pro-
vided at Cratylus 395ci (kata to atreston), Atreus' name reflects his lack of fear—
a derivation whose correctness is underscored by Socrates (c2). As Sansone (1996,
62) observes, Plato's source here may be Euripides' Iphigenia in Aulis, where
Agamemnon says, with himself in mind, "I am the son of Atreus. Do you think he
shrinks from your eye, Menelaus?"
321). As previously observed, this comment, like authors' treat-
ment of Hector and his son, reflects the assumption that a son's nature—at least
when judged or imagined to be exemplary—will mirror that of his father.

34. On her role as avenger, see Aesch. Cho. 948-51, Sept. 662-73; Hes. Op. 213-73; Soph. A/.
1389-92.
 ETYMOLOGY AND THE CRATYLUS SOURCES 57

Plato's analysis of "Agamemnon" ("Admirable-for-remaining") emphasizes

the hero's steadfastness (Cra. 39535-7). On his interpretation, moreover, the
name is properly assigned since its descriptive content succeeds in capturing
Agamemnon's nature (39535^3). Euripides, in contrast, may imply the name's
inappropriateness in Iphigenia in Aulis, where Menelaus is sharply critical of
his brother for lacking resolve:

You're crooked; always changing—

this way, that way, back again. . . .
Your shiftiness is a doubtful asset to your friends. . . .
A man of principle should not change character
as he grows great. When good fortune gives him power to
help his friends, that's the time when
most of all he ought to prove reliable.
There's my first criticism, the point where I first found you treacherous. . . .
Then, when Calchas bade you offer sacrifice to Artemis
with your daughter's life, and promised us fair winds, then
you were pleased,
gladly undertook to kill her. . . .
Now you change your mind. (332-63)

Regarding the possibility that a judgment of inappropriateness is suggested in

this passage, it is noteworthy that Menelaus cites repeatedly his brother's changes
and changeability, making emphatic use ofbebaios ("firm," "steadfast") to under-
score what Agamemnon is not (334, 347)-
As I have argued, Plato's recourse to the literary tradition when exploring the
descriptive content ofonomata is strikingly evident in the Cratylus. Interestingly,
the Cratylus is not the only place where Plato attends to the way in which names
and their semantic constitution are exploited in a literary context. When Plato
treats the topic of early education in Republic 2-3, he has much to say about the
content of acceptable poetry. One stricture on which he insists is that poets'
constructions not include material that fosters undesirable emotional responses,
prominent among them fear and terror (3863-3870). Of special import vis-a-vis
the Cratylus is the fact that Plato introduces explicitly the topic of names:

We must do away with all those frightening and terrifying names that crop up
here, names like Cocytus and Styx, ghost and wraith, and so on —all those names,
in fact, that are designed to make everyone who hears them shudder (KCOKWOXX;

Though they may be

of some use in another context, our worry is that this shivering might make our
guardians too feverish and cowardly. (387b8-c5)35

35. I follow closely here the excellent rendering of Waterfield (1993) and endorse his omis-
sion of obi; o'lETca in C2. Literature supplies good evidence of these names' effectiveness, as when
58 THE ANCIENT QUARREL

Whereas onomata with negative descriptive content, namely, those promoting

undesirable attitudes and emotional responses, will be excluded from poetic
compositions that are to be acceptable pedagogical tools, "names producing the
opposite effect must be included in both prose and verse"
v, 38709). Thus, Plato precludes the use
whose impact is deleterious, yet wishes to harness the power of names where it
can aid in the promotion of his own educational goals.
In addition, early in the Republic's discussion of acceptable poetry, Plato
contends that

god, since he is good, could not be responsible for everything

as the populace asserts. Rather, for human beings he is the cause of few things,
but of many not the cause. For, in our case, good things are far fewer than bad,
and the good things we must assume to be the sole responsibility of god, while
for the bad we must seek explanatory factors other than god

(jygc)

Plato's first stipulation regarding acceptable poetic compositions treating the gods
is thus that those entities be depicted invariably as "responsible, not for all things,
but solely for the good" (UT) Trdvioov amov . . . dtAla TCQV dya6a>v, 380^8-9).36
Based on his subsequent remarks about names and their descriptive content
(jSyb-c), one would expect Plato to favor analyses of familiar deities' onomata
that did not allow a harmful impression of the nature of divinity to form.
Taking this important discussion in the Republic as a reference point, let us
return to those derivations in the Cratylus that concern individual deities' na-
tures and have literary counterparts. If one compares the two sets of etymologies
in light of the criteria articulated in the Republic, Plato's analyses may be seen
to follow a striking pattern.
Where the inherited analysis is neutral or positive— from that perspective
endorsed in Republic 2-3 — Plato follows the derivations offered by his prede-
cessors. Thus, for example, Plato is content to embrace Euripides' analysis of
Demeter's name (Hel. 1301-2, 1320, 1327-28, 1340, 1356; Phoen. 683-86; Bacch.
274-76), which emphasizes her role as nurturer of humanity. He also retains

Hesiod refers to "fearsome Styx, a goddess loathsome to the immortals"

Th. 775-76; my translation). See also Circe's foretelling that Odysseus will arrive
finally at "the crumbling homes of Death. Here, toward the Sorrowing Water, run the streams of
Wailing, out of Styx, and quenchless Burning
(Horn. Od. 10.512-14). Regarding the Furie
whose treatment, as we saw in chapter i, involves a name change to "Eumenides" for apotropaic
purposes, see Soph. OC 39-43, 486-87.
36. Cf. 3776-3783, where, having dubbed tales of the gods' taking revenge on one another "lies,"
Plato then claims that, even if such tales were true, they should not be generally circulated.
 ETYMOLOGY AND THE CRATYLUS SOURCES 59

Homer's stress on Athena's intellectual prowess. These analyses are advisedly

preserved, in light of Plato's strictures at Republic 387!)-^ due to their positive
content and hence impact on their auditors. Moreover, in the Cratylus, Plato
has a choice between two literary derivations of the name "Aphrodite" and opts
explicitly for Hesiod's analysis, which, as we have seen, traces the goddess's name
to aphros ("foam"). Since Euripides' derivation, in contrast, makes a serious
character flaw (aphrosune, "folly") integral to her nature, it would be objection-
able from the Republic's perspective. It is interesting to note in this connection
the phrasing of Plato's endorsement of the former semantic breakdown: "In the
case of Aphrodite, it is not worthwhile to challenge Hesiod
(40607).37 It is arguably the etymology's neutrality from the van-
tage point of those criteria set forth in the Republic that obviates the need for a
confrontation. Finally, Homer's derivation of "Hera" from "air" is also morally
neutral, and metaphysically benign or even positive (insofar as it indicates divine
governance of the cosmos); hence, it too may be preserved without risk of harm.
Where, however, as is more frequently the case, literary etymologies violate
the Republic's criteria, Plato revises the derivations found in literature. Some-
times the distinction between retention and reconstitution, while crucial, is
subtle, as in the pivotal case of Zeus. In literary etymologies, Zeus is described
in precisely the inconsistent way that Plato sharply criticizes in the Republic. As
he stresses there (3790-6), this god above all must be depicted as responsible,
not for both good and evil, but rather solely for the former; Plato's examples of
passages to be stricken thus include Homer's comment that "there are two urns
that stand on the door-sill of Zeus. They are unlike for the gifts they bestow: an
urn of evils, an urn of blessings" (II. 24.527-28). Plato's procedure in the Cratylus,
when understood against its literary backdrop, suggests how he would handle
the etymological aspect of the problem. Zeus's name was an especially popular
object of etymological analysis —not surprisingly, given his status as supreme
Olympian.38 As we have seen, literary derivations trace the two forms of the god's
name to zen and dia. While an author's account might underscore a positive
effect or exercise of Zeus's power (see Apollo's comment in Orestes), one may
also find the responsibility for evil placed squarely at his door, as when the Cho-
rus in Agamemnon depicts the god as ultimately responsible for the disasters of
the House of Atreus (1485-87). This attribution of a negative causal role to the

37. Insofar as Plato fails to endorse the derivation from aphrosune, one might wish to view
his reaction as involving revision. I consider a necessary condition for the presence of revision to
be that Plato act on inherited material in a new way; because he endorses expressly an analysis
provided by a figure in the literary tradition, I judge his handling of "Aphrodite" to involve reten-
tion. This criterion serves to distinguish the case of Aphrodite from that of Zeus, to be discussed
below.
38. Underscoring Zeus's unique status among the gods, Burkert (1985,131) notes that he "was
the only god who could become an all-embracing god of the universe. The tragedians did not
present him on stage, in contrast to Athena, Apollo, Artemis, Aphrodite, Hera, and Dionysos."
6O THE ANCIENT QUARREL

god, which is part and parcel of Aeschylus' etymology, is explicitly and strenuously
prohibited by the Republic. Plato's consolidation of the analyses involving zen and
diet into a single derivation—which makes Zeus the one "through whom all crea-
tures always have life" (Cm. 396^) —thus has a twofold function. First, as we saw,
it shows that he can top the poets' performance on the same linguistic terrain. More
important from a philosophical perspective, while Plato works with the same two
source onomata as literary derivations, his consolidation of them represents a cru-
cial revision of literary practice. The key innovation here is that Plato's analysis,
unlike those of the poets, stipulates what the god is responsible for: life itself, and
only that. This move allows Plato to block causal attributions to the god of other
phenomena —including evils such as befell the House of Atreus—for which the
god might otherwise be deemed accountable. Since etymologies and depictions
of the gods' characteristic activities are often tightly interwoven, it becomes cru-
cial for names' semantic constitution to convey the right message. Early influences
on the young are pivotal factors shaping the ways in and extent to which individu-
als realize their natures as human beings, and whatever opinions (doxai) are ab-
sorbed at this juncture typically become permanent fixtures of individual psyches
(Rep. 3773-c, jySd-e). Thus, if the import of names is not the same as, or at least
consistent with, that salutary message conveyed by other means, Plato's central
pedagogical project of inculcating the proper attitudes and emotional responses
in the young risks being utterly compromised.
Literary analyses of "Ares" (from ara) and "Cronus" (from kraino, but where
ara too appears) emphasize the gods' role as the source of curses, which were
not infrequently identified in literature as a means of achieving revenge. Nota-
bly, at Republic jyye^ySb—a passage in which Cronus himself is mentioned —
Plato insists that revenge not be depicted as a divine occupation. Interestingly,
when Plato articulates the gods' natures via etymology in the Cratylus, in nei-
ther case does his account privilege the activity of cursing. Plato's own deriva-
tion of "Cronus" —it signifies koros, "as in 'the pure and unmixed character of
the mind itself" (396b6-y) — instantiates his frequent preference for analyses that
emphasize the moral or intellectual assets of names' referents. This predilection
is also evident in Plato's treatment of the name "Ares," which includes a men-
tion of its referent's bravery (to andreion, 407^2). Literary analyses of "Dike," in
turn, emphasize the deity's position as the daughter of Zeus; most central here
is the fact that requests for vengeance figure prominently in poets' characteriza-
tions of her (see Hes. Op. 213-73; Aesch. Cho. 948-51, Sept. 662-73; Soph. A/'.
1389-92). Plato's treatment of the relevant terms (dikaiosune and dikaion, 4120-
4i3d), as we saw, moves in quite a different direction.
Plato stresses the fear generated by the names "Hades" and "Apollo" due to
inherited analyses.39 His strategy here is, first, to emphasize the existence of these

39- F°r Plato's emphasis on the fear elicited by names, see also his treatment of "Persephc
(404c-d).
 ETYMOLOGY AND THE CRATYLUS SOURCES 6l

derivations and their impact on mortals, then to derive the names from differ-
ent sources so as to cancel out or erase this negative capability. Mortals' fear of
Hades, which stems from the name's being traced to aides, leads them to call
the god "Pluto" instead (4033 [cf. 4O2d8]; for examples of this substitution, see
Aesch. PV 806; Eur. Ale. 360, HF 808). This euphemistic onoma allots to its
referent a role in the giving of the wealth 40334) that
issues from the earth, hence its outcome with respect to the god is the substitu-
tion of a good function for bad. There was, one may recall, a separate deity named
"Plutus" who was the son of Demeter and lasion (see Hes. Th. 969-
71; Hymn to Demeter 489).40 On the foregoing literary and Platonic construc-
tion, Hades is thus made to share his function. Also prompted by fesr, snd eu-
phemistic, is the substitution for "Hades" of alternative phrases that include the
term "name" in one form or another, as in "the many-named son of Cronus"
Hymn to Demeter 18). Plato's use of the locution
"the other name" thus recalls through instantiation
this second technique.
From Plato's perspective, the alleged "solution" described above is thoroughly
inadequate becsuse it is itself entirely governed by that from which it is designed
to escape. If anything, promoting linguistic replacements will intensify mortals'
fesr by serving as an eloquent reminder that the deity in question is one whose
destructive nature and hold over them are so pronounced as to require an ex-
treme response. A more satisfactory resolution of the problem would consist in
the provision of an alternative breakdown of "Hades" that invested the onoma
with positive descriptive content. Plato's own alternative (404!)) highlights the
god's superlative intellect: far from the name's being connected to the "unseen"
(aides), it is derived from the god's knowledge of all admirable things
In the Republic, Plato underscores the paramoun
concern of early education with the fostering of the right attitudes toward the
admirable (kalon) and shameful (aischron) (4016-4023). From the vantage point
of that discussion, the name "Hades," as construed 3t Cratylus 404^ would be a
useful pedagogical tool insofar 35 it stressed the existence of a divine concern with —
and hence the prsiseworthiness of—whst is fine as opposed to bsse.41

40. Citations of the Hymn to Demeter are from the edition of Richardson (1974).
41. That Plato finds this etymology of "Hades" more promising from a philosophical stand-
point is also recognized by Barney (1998, 69-70) and Friedlander (1964, 207-8), who do not,
however, note a connection to Greek literature or the Republic. Interestingly, Plato offers else-
where an elevated interpretation of that Homeric derivation which the Cratylus rejects when lie
says that after death the invisible part of the soul goes off to the realm of Forms, that is, "to Hades
in the true sense of that term" . At most, this revised con-
struction of the traditional derivation reinforces a point made on independent philosophical
grounds and as such plays no essential role in the discussion. In both the Cratylus and the Phaedo,
Plato's choice of derivation is thus based on the requirements of that context in which it is to play
a role. Fowler (1926, 4) suggests that Plato's rejection in the Cratylus of an analysis of "Hades"
which he accepts at Phd. Sod "may indicate that the [former] is the later of the two dialogues."
62 THE ANCIENT QUARREL

The locus classicus in Greek literature for the view of Apollo as destroyer is
Cassandra's lament at Agamemnon 1080-82: "Apollo, Apollo! Lord of the ways,
my destroyer. You have undone me once again, and utterly." Acknowledging
the strong reaction generated by his onoma, Plato observes (Cra. 40461-2) that
"many people are seized with fear regarding the god's name because they be-
lieve that it reveals something dreadful (for an em-
phasis on mortals' fear, see also 4O4C5~y and 40562-40631). Far from endorsing
this negative view of the god, Plato concludes that people have serious miscon-
ceptions about his nature, which he attributes to their mistaken beliefs about
the descriptive content of his name. A proper etymology should therefore sup-
plant false opinions with true by replacing the offending source onoma —in this
case the verb apollumi.*2 Having underscored the name's familiar negative im-
pact, and his sharp divergence from this interpretation, Plato offers, as we saw,
not one, but four alternative derivations. These, taken together, bring out what
he identifies as the god's salient positive capacities (4O5d-4o6a). Interestingly,
in the Republic (3833-0), Plato rejects Aeschylus' description of Apollo (frag.
35oNz) as the destroyer of Achilles: we will be furious, he asserts, with any poet
who says this type of thing about the gods and will not permit such material to
be used in education if the young are themselves to become divine to the extent
possible for human beings. Plato's fourfold analysis in the Cratylus constitutes
one prime, if ambitious, example of the sort of linguistic elevation of which the
Republic would doubtless approve.
On the foregoing account, the Cratylus itself offers numerous examples of
the correct literary practice sketched in Republic 2-3, in particular of the results
of the remedial work on names involved therein. Where names' referents are
traditional gods, it would be a problematic move indeed to replace these onomata
with others. And in fact this extreme measure is nowhere advocated by Plato,
who is concerned instead with shaping the right conception of their referents'
"natures." To promote this end, familiar divine names must be traced where
necessary to different sources—in particular, to terms whose referents are more
positive in Plato's conception of moral and intellectual qualities—with the re-
sult that audiences' views of those entities' natures will be molded accordingly.
Although from his perspective these entities do not have natures (phuseis) in the

In my view, such an approach, which presupposes that Plato took individual derivations seriously
in their own right and ignores the importance of context, places more weight on particular analy-
ses than Plato's etymologizing in the Cratylus and elsewhere can sustain. For further discussion
of Plato's use of etymology outside the Cratylus, see chapter 3, section 3.
42. Cf. Baxter (1992,145), who notes that one may plausibly infer from this rejection olapollumi
that "Plato calls into question the traditional picture of Apollo that emerges from myth and trag-
edy," adding that "here one can compare Plato's strictures in the Republic on the allegorical inter-
pretation of the traditional stories of the gods: it is wrong and educationally damaging that the
gods should be represented as quarrelling, wronging each other and so forth."
 ETYMOLOGY AND THE CRATYLUS SOURCES 63

primary sense of that term, phuseis proper (namely, Forms) can be grasped only
via reason, which is avowedly not, indeed cannot be, the direct target of early
education (4016-4023).4? While pedagogy does not, at this stage, itself target the
intellect and hence does not introduce natures proper into the curriculum, its
ultimate goal is to facilitate the understanding of Forms (above all, those per-
taining to values and mathematics) in adults whose own natures suit them to its
attainment. From the perspective of the Cratylus, familiar literary practice may
be faulted for operating with a misguided view of what, first and foremost, is or
has a nature and in the belief that etymologies can illuminate reality itself. The
existence of a consistent pattern in the dialogue's retention and revision of liter-
ary etymologies of divine names is intriguing insofar as it may be taken to sug-
gest that this tradition also fails to grasp adequately even those secondary natures,
or "divinities," that are pivotal to its own constructions and are to figure promi-
nently in the poetry incorporated in Plato's ideal community.44
Regarding the issue of ties between Platonic and literary etymology, one may
note, finally, that the terminology used in the Cratylus to treat questions of
appropriateness—principally the adverbs orthos, alethos, dikaios, and kalos ~par-
allels in striking ways that employed by the literary tradition. This is especially
true of orthos, as used, for instance, in Aeschylus' Seven against Thebes (829),
Euripides' Alcestis (636) and Trojan Women (990), and Herodotus (4.59, orthotata).
Alethos is found, for example, in Aeschylus' Suppliants (315) and Euripides'
Phoenician Women (636). Terms with the same root as dikaios appear in Aeschylus'
Seven against Thebes (670, pandikos) and in Euripides' Iphigenia in Tauris (500,
dikaion as adverbial accusative) and Alcestis (647, endikos). Kalos is found in
Aeschylus' Libation Bearers (951).45 The terminology employed by Plato to dis-
cuss orthotes onomaton would appear to have strong normative connotations.
Since he regularly invokes the concept oftechne in discussions of activities whose
performance involves expertise of an elevated sort, the fact that Plato uses terms
like orthos and dikaios in the Cratylus' handling of fitness prompts one to con-
sider whether his doing so indicates a belief that naming qua etymologizing
qualifies as a practice of this caliber. It is to this issue, along with others involv-
ing the Cratylus' philosophical concerns and outcome, that the following chapter
turns. Before we take up these matters, however, it is necessary to round out the
discussion of Plato's sources by commenting on other candidates whom inter-
preters have proposed as foundational to his treatment of etymologizing.

43. Alternatively, one might make the point in terms of the "divine," the idea being that Forms,
not the Olympian gods, bear divinity's true marks; on this issue, see p. 96 with n. 43.
44. See chapter 5 for a defense of the claim that poetry does have a role, albeit a carefully
circumscribed one, in Plato's Republic.
45. In the Cratylus, see, e.g., orthos: 39262, 39469, 395b3, C2, d4, 39639, 03, 6, 397by, c6, 39803,
39905, 40004, 40107, 4O4d4, 40501, 40601, 40735, 40906, 41202, 41335, 4^4, 62, 4i6d8; dikaios:
40606 (dikaiotat'), 4o8bi, 40901 (dikaiotaf), 4iob7, 09, 41802, 4igd7 (to dikaion); alethos: 39762,
4oob6; and kalos: 39od6, 39631 (pankalos), b8, 4oobi, 40^7.
64 THE ANCIENT QUARREL

3. Other Proposed Sources of the

Cratylus' Discussion of Etymology
As noted, the foregoing argument does not aim to show that the literary tradi-
tion is the only source of the etymological practice that is the centerpiece of the
Cratylus. Rather, it introduces considerations in support of the view that, based
on extant evidence, the literary tradition is the most direct and prominent oppo-
nent of Plato's critique of etymology. Since earlier discussions have devoted
comparatively little concern to the literary tradition, this study concentrates on
illuminating its importance and gives only limited attention to more familiar
proposed sources. The preceding account indicates why interpretations giving a
pivotal role to Prodicus and Protagoras are not convincing. In what follows, I
comment on a range of other candidates, suggesting why discussions that assign
them a central place are also open to question. Since many such challenges have
already been posed by others, where pertinent I refer the reader to discussions
in which salient considerations are introduced.
Virtually nothing is known about Cratylus other than what is said about him in
Plato's dialogue. According to Steinthal ([1890] 1961,82), one can use the dialogue
as a source of information about the man behind the figure of Cratylus; it is "en-
tirely probable," he contends, that Cratylus had turned Plato's attention toward
etymology before Plato studied with Socrates (83). Wilamowitz (1959, 223) asserts
confidently that Plato derived an interest in etymological practice from his asso-
ciation with Cratylus. In the view of Rijlaarsdam (1978,10), "what Plato derides
in the etymological section is the standpoint of Cratylus, according to which
'etymologizing' and 'philosophizing' are the same, and one can therefore gain
access to the essence of things through the analysis of names"; despite this criti-
cal stance, "Cratylus' authority as an etymologist was so great that even Plato
viewed his interpretations of names as probable" (183).46 Derbolav (1972, 29-30
[with n. 5]) and Meridier (1931, 40) have noted various difficulties involved in
correlating the historical Cratylus with that stance to which his name is linked
in Plato's dialogue. Although, as Meridier points out, the dialogue's title might
seem to support the view that Cratylus is Plato's target, "nothing allows one to
affirm that Cratylus, so poorly known from other sources, engaged in etymolo-
gizing"; what is more, his conduct in the dialogue "gives the opposite impres-
sion." Meridier cites in this connection, for example, the fact that "Cratylus
neither proposes nor discusses any etymology; he limits himself to accepting
docilely the whole group of those offered by Socrates, and if he hastens to ap-
prove them, it is because he finds in certain ones confirmation of the flux-doctrine
of Heraclitus." A recognition of Cratylus' absence during the framing and con-
duct of the etymological debate leads Heath (1888,198) to make the following

46. Regarding the pertinence of Cratylus, see also Goldschmidt 1940, 93; Barwick 1957, 74;
and, most recently, Sedlcy 1998, 146.
 ETYMOLOGY AND THE CRATYLUS SOURCES 65

comment: "What kind of theory Cratylus would have propounded had he

roused himself to the effort without the intervention of Socrates one does not
know."
Casting a wider net, Jowett (1892, vol. i, 253) claims that "had the treatise of
Antisthenes upon words, or the speculations of Cratylus, or some other Hera-
cleitean of the fourth century B.C., on the nature of language been preserved to
us; or if we had lived at the time, and been 'rich enough to attend the fifty-
drachma course of Prodicus' (384 B), we should have understood Plato better,
and many points. . . would have been found . . . to have gone home to the soph-
ists and grammarians of the day." Others too mention Antisthenes. Steiner (1916,
127-29), for instance, contends that he is the target of the etymologies in 4113-
4210. Guthrie (1971) admits that it is hard to be sure what Antisthenes' teaching
was (209), yet asserts that it resembles Cratylus' natural-correctness thesis (215)
and that both Antisthenes and Plato's Cratylus believed onomata to have a "natu-
ral affinity" with their objects (219). Not discovering a substantive connection
between any notion of natural affinity that might be attributed to Antisthenes
and the particular form that concern takes in the Cratylus, Steinthal ([1890] 1961,
122-23) "ghtly rejects what he deems the speculative view that Antisthenes fo-
cused on etymologizing.47
While Antisthenes is said to have had some interest in onomata*8 very little
is known about what his view actually was. On the assumption that a connec-
tion between names and education is correctly attributed to him, there is no
evidence that his more specific position is like that which governs the etymo-
logical section of the Cratylus. What is more, there are indications that his view
of naming and natures would have diverged from that investigated by Plato. For
one thing, Antisthenes' view of elements and what can be said thereof (on which
see, e.g., Arist. Metaph. io43b23~32; Steinthal [1890] 1961, 123) does not seem
compatible with the Cratylus' idea thatonomdta are not basic (i.e., unanalyzable)
units but instead — if properly traced to other linguistic items—themselves sources
of insight into their referents' natures. With respect to the topic of natures,
Antisthenes is alleged to have claimed that while, according to convention, there
are many deities, by nature there is only one.49 This position concerning natures
is at odds with that explored in the Cratylus, according to which it is assumed
that etymological analyses of divine onomata, if properly carried out, disclose

47. For challenges to the view that the Cratylus targets him, see also Meridier 1931, 44~45;
Taylor 1960, 86 n. i, 89 n. i; Wilamowitz 1959, 223, 230; Levinson 1957, 31-32. Cf. Derbolav 1972,
31 n. 6: "Circular appears to us ... to be an argument that, if it does not derive its picture of
Antisthenes from anonymous Platonic doctrines, nevertheless enriches it thereby and with it dem-
onstrates the identity of Cratylus with Antisthenes."
48. A work with the title is attributed to him (Diogenes Laertius
6.17), as is the statement (Caizzi 1966, frag. 3
Steinthal ([1890] 1961, 124) expresses doubts about the latter attribution.
49. Caizzi 1966, frag. 3gA.
66 THE ANCIENT QUARREL

their individual referents' phuseis. Moreover, Antisthenes is also supposed to

have claimed that since god (theos) is not like anything else, one cannot learn
about this being from any sort of image (ef&on). 50 If onomata are encompassed
by the term eikon, it would follow that etymologizing will not, in Antisthenes'
view, yield insight into the nature of the one divinity. Again, such an assump-
tion would be at odds with that underlying the practice investigated in the
Cratylus, according to which etymologies do provide valuable guidance regarding
deities' natures. Moreover, and quite importantly, the Cratylus of Plato's dia-
logue is strongly attached to the view that names are eikones of what they denote
(431c-433c).
Warburg (1929) argues that the Cratylus is directed against Heraclides Ponticus,
the fourth-century Academic philosopher, whom Warburg maintains combined
etymologizing and Heracliteanism. 51 In his view, "among the etymologies trans-
mitted under the name of Heraclides there are some that are in agreement, not
only methodologically, but also substantially and literally with their counterparts
in the Cratylus" (25-26). 5Z Yet, as von Arnim (1929, 22) has observed, rather than
discussing all parallel instances that he alleges to have discovered, Warburg treats
only one case, the etymologies ofselene in the Cratylus and Orion, concluding
(Warburg 1929, 27) that Plato's derivation is related to Heraclides' "as variation
to theme, that is to say, as something secondary to what is primary." 53 Such evi-
dence is far too slim to serve as a foundation for Warburg's claim. In addition,
Dahlmann (1929, 1674) maintains that the material in Orion to which Warburg
appeals in support of his theory suggests that the Heraclides mentioned therein
is an altogether different individual. Wehrli (1969, 117-19) has argued that Orion's
source is in fact Heraclides Ponticus the Younger (a grammarian and student of
Didymus), who was active during the first century A.D. More recently, Gottschalk
(1980, 160-62) diverges somewhat from Wehrli on the level of specifics but offers
further support, from a different angle, to the view that the Academic Heraclides
is not the backdrop for Plato's Cratylus. The evidence provided by the foregoing
scholars, taken together, seems to me to be decisive.
At Cratylus 3g6d, Socrates claims to have taken his inspiration from Euthyphro.
While references such as this have led some to conclude that the historical fig-
ure is a key backdrop for the etymological discussion, in particular that section
involving divine onomata,^ as Baxter (1992, 108-13) has argued persuasively, the
Cratylus does not support the view that Euthyphro has this role.

50. Caizzi 1966, frag. 40.

51. On the account of Gottschalk (1980, 5), Heraclides was an active member of the Acad-
emy at least from the latter half of the j6os until 339.
52. Cf. Thesleff (1982,168-69), whose attempted defense of Warburg is, as Baxter notes (1992,
95 n. 59), unpersuasive.
53. For von Arnim's challenge to Warburg's thesis, see more generally 20-26.
54. See Barwick 1957, 73-74; Steiner 1916, 124-26. On Euthyphro's pertinence, cf. Levinson
1957, 36.
 ETYMOLOGY AND THE CRATYLUS SOURCES 67

As Kahn (1997, 55) observes, nothing is known about Euthyphro aside from
the information provided by Plato, and his name appears only in the Euthyphro
and Cratylus.^ In the former dialogue, Euthyphro is remarkably arrogant and
emphasizes repeatedly his access to wisdom not available to ordinary people.
There is no hint that this alleged wisdom is based on or otherwise linked to ety-
mology; the idea is simply that special ties to the divine realm issue in privileged
access to important truths. There are various reasons why Plato might introduce
such a character into the Cratylus' discussion. The possibilities include a wish
to distance himself from the inquiry into orthotes in the form it takes in that
dialogue;56 an interest in dissipating any element of strangeness associated with
the discussion's having Socrates as its leader;57 and a desire to account for how
Socrates could apprehend the esoteric matters of which he speaks. While one
cannot be sure which, if any, of these grounds is most prominent, the third pro-
posal finds some support in the fact that it meshes with the picture of Euthyphro
gained from the earlier dialogue.58 Some privileged understanding would indeed
seem to undergird the production of that lengthy stream of derivations offered
by Socrates. Additionally, in the Cratylus' discussion, Socrates makes a range of
observations (e.g., about words' being highly compressed versions of lengthy
phrases and about their having ancient or foreign origins) that would appear to
testify to his possession of special insight. Cratylus' reference at 42807-8 to
Euthyphro "or some other Muse" suggests that he as an individual is not what is
important from Plato's point of view.59 Moreover, this particular formulation is
noteworthy since it points to an analogy with the inspiration of poets and hence
a tie to the literary tradition (see, e.g., Horn. II. r.r, 2.484, Od. 1.1-10; Hes. Th. i,
Op. i); interestingly, Cratylus quotes Homer just prior to his mention of Euthyphro
(42804-5), and a little earlier (ai) Hermogenes mentions Hesiod. For a related
passage, one may consult Euthydemus 27jc-d, where Socrates says that he must,
like the poets, invoke the Muses for help in completing the difficult task at hand.
When addressing the issue of individuals' status as targets of the Cratylus'
discussion, one must distinguish between Plato's taking account of others' views
by way of etymology and his targeting them specifically for their use of this device.
Plato's etymologizing offers clear support to the view that he engages the liter-

55. Although one cannot prove beyond all doubt that the character in both places represents
the same historical individual, the conclusion is quite plausible and will be assumed here; for
justification of this view, see Baxter 1992, 108.
56. For emphasis on a motive involving detachment, one may consult the claim of Barney
(1998, 73) that "the real function of the genre of inspiration is ... to distance the contents pre-
sented from the figure of Socrates" (italics in original).
57. On this point, see Gaiser 1974, 19.
58. Such proposed grounds need not, of course, be mutually exclusive; my concern is to iden-
tify the factor that has the strongest textual support.
59. Cf. Meridier's appeal to this passage (ig^r, 42) to support the claim that "this mediocre
seer and those in his circle are not the true object of his attacks."
68 THE ANCIENT QUARREL

ary tradition on both of these levels. Since I have already discussed this tradition's
role on the latter plane, I note here that Plato engages it also in the former regard,
as evidenced, for example, by his analysis of herds based on heroes' mode of
genesis and role in literature as skillful speakers (398c-e); his first etymology of
psuche, which draws on the idea, prominent in Homer, that the soul is respon-
sible for life (4<Doa);60 and the fact that his account of how Hesiod would ety-
mologize the term daimon has as its point of departure the poet's description of
its referent (3976—3980). Moreover, as we saw, the pattern discernible in Plato's
revised derivations of individual deities' names suggests a critical stance toward
literary ideas about divinity. Evidence is lacking, in contrast, for the claim that
Plato takes substantial account of both factors in the case of Anaxagoras, Democ-
ritus, and Heraclitus, supporting instead the view that the Cratylus' engagement
with them pertains largely to the positions they espouse.
While the fact that Anaxagoras etymologizes aither (see Arist. Gael. 270^24-
25) constitutes one link with the Cratylus on the etymological plane (though
not in the content of the proffered analyses), there is no evidence in the surviv-
ing fragments of a substantial use of this device to illuminate natures. What is
clearly in evidence, however, is Plato's engagement in the Cratylus, as in the
Phaedo, with Anaxagoras' view of reality. Thus, Plato's second etymology of
psuche stresses the role as ordering principle given by Anaxagoras to mind and
soul (4<Doa-b), and his derivation ofselene purports to show that Anaxagoras did
not originate the view that the moon receives its light from the sun (see Cra.
4C>9a-c and DK 18).61 Still later, Plato introduces Anaxagoras' conception of mind
in his treatment of dikaion (4133^).62 As Baxter (1992, 128-29) points out, Plato
stresses here the limited efficacy of etymology: "even if everyone is agreed that
8i,Kcaov is to be derived from 8id iov, there is nothing in the etymology itself which
especially recommends votx; as what 'goes through'; it is an answer that will only
convince Anaxagoreans." In assessing Cratylus 413, one cannot but think of Plato's
critique of Anaxagoras and other Presocratic philosophers of nature in the Phaedo,
particularly given the fact that in both dialogues Plato is concerned with what
constitutes a genuine, rather than a merely specious, explainer (to aition).6^

60. On psuche in Homer as "the force which keeps the human being alive," see Snell [1953]
1982, 8. The term psuche is used there only in what Nussbaum (1972, 3) calls '"negative contexts,'"
namely, when its continued presence in the body is threatened or its ties thereto have already
been severed.
61. A tie between mind and soul was certainly envisaged
DK 12). . As Aristotle observes, however (DA 404^-6, 405313-19), Anaxagoras was not as
clear as he might have been regarding this connection.
62. Concerning Anaxagoras' view of mind, see DK 11-14. Aristotle underscores its special sta-
tus for Anaxagoras as the arche of everything, alone in being simple (haploun), unmixed (amige),
and pure (katharon) (DA 405315-17).
63. For this term, see Cra. 41334-5; at Phd. ggb, Plato distinguishes between a genuine explainer
and a mere necessary condition, or "that without which the explainer (TO auiov) would not be an
explainer (aluov)." Regarding Anaxagoras' depiction in the Phaedo, cf. Baxter 1992, 129.
 ETYMOLOGY AND THE CRATYLUS SOURCES 69

According to Baxter (1992, 157), in the Cratylus "Democritus' theories are

being gently mocked using one of his own tricks, etymology." In his account of
Democritus, however (156-60), Baxter does not always distinguish sufficiently
between Plato's possible critique via etymology of Democritus' ideas and his
engaging Democritus with a view toward his use of etymology per se. In fact,
the evidence offered by Baxter pertains mainly to the former. Thus, he cites
hekousion ("voluntary") and ananke ("necessity"), whose etymologies (4206-6),
Sambursky (1959) argues, link Democritean and Pythagorean ideas. Insofar as
Democritus is relevant to the derivation ofsophia (4izb), which emphasizes touch
(ephaptesthai), his influence here too is described by Baxter (156-57) as operat-
ing on the level of doctrine. Based on this construction otsophia, Baxter specu-
lates, in addition, that Plato's etymology of phronesis (4116) "is referring to the
same Democritean ideas" (157). With regard to etymology itself, the strongest
connection is between Cratylus 4143 (gune-gone) and DK1223. DK 2 (Tritogeneia
he Athena) identifies Athena with phronesis because to phronein admits a three-
fold division. Allegory and etymology may, but need not, be combined; although
one could translate the foregoing observation about Athena into an etymology
of her name, Democritus himself does not undertake this. Hesiod too calls her
"Tritogeneia," in a myth (T/z. 886-98) that "allegorizes the three basic charac-
teristics of. . . Athena: her prowess, her wisdom, and the masculinity of her vir-
gin nature" (Morford and Lenardon 1995,117-18); in addition, as we saw, Homer
interprets the name "Athena" itself in terms of the goddess's intelligence.
Regarding Democritus' possible role as backdrop for Plato's discussion, it is cru-
cial to note, moreover, that while the words agalmata phoneenta (DK 142) might
appear to be suggestive, the fragment explicitly distinguishes this view of the gods'
onomata, attributed to Democritus, from the Cratylus' construction of natural
appropriateness (regarding the phrase's interpretation, see Guthrie 1965,475-76).M
In the view of Kahn (1973,155-56), "if we had more information on Democritus'
views, there are probably many puzzling passages in the Cratylus which would
appear in a clearer light." While one cannot deny that this is so, the fact remains
that based on surviving evidence a significant tie cannot be forged between him
and Plato regarding etymology specifically. According to Proclus (DK 26), in
Democritus' view onomata are correct by convention, not nature
in this same fragment, Democritus, grouped with Aristotle,
is said to accord with Hermogenes. This fact alone would arguably suffice to exclude
Democritus as a central target of the Cratylus' natural-correctness thesis.
The Cratylus gives an important role to the doctrine of flux, to whose alleged
preeminence a large number of etymologies are made to testify, thus making

64. On the account of Hirschle (1979), it is quite probable that the view of names as agalmata
was not actually that of Democritus but came to be linked with him due to an error committed
in haste by the student who took down the notes that have come down to us as Damascius' com-
mentary on the Philebus. For the details of this argument, see Hirschle 1979, 63-65.
70 THE ANCIENT QUARREL

Heraclitus a noteworthy Platonic opponent on the level of ontology.65 His pro-

posed influence has been extended to the realm of naming by Marcovich (1967,
193), who suggests that "as far as the is concerned,
Heraclitus might have been the spiritual father of Plato's Cratylus."66
Having challenged the view that this figure may be assumed to represent
Heraclitus' own stance, Kirk (1962,119) asserts that "a much more cogent indi-
cation of the fact that for Heraclitus names bore some essential relation to ob-
jects, and were capable of revealing a truth about them which might not be other-
wise obvious, is provided by the not uncommon instances of etymology in the
tragedians and especially in Aeschylus." One cannot simply appeal to poets to
ground the contention that Heraclitus held a particular view of the name-thing
relation. Moreover, from the claim that Heraclitus and the tragedians believed
in "some essential relation" between the two, it need not follow that this rela-
tion took just the same form for Heraclitus and the poets. Indeed, Kirk notes
that while "for Heraclitus there is a real and essential connexion between the
name and the thing named, . . . the nature of this connexion is not explored"
(198). The following remarks offer support to the claim that Heraclitus did not
investigate the ties in question. What remains unclear is whether this lack of ex-
ploration is simply a misleading impression generated by the fragmentary nature
of surviving evidence, or whether Heraclitus himself was simply not interested in
pressing the issue further than he does in the material available to us.
In modern discussions of ancient sources, terms like "wordplay" and "pun"
are used to encompass numerous phenomena, and a failure to distinguish
adequately between types of wordplay or punning has sometimes led commen-
tators to posit closer ties between different authors' efforts than is justifiable based
on the particular subject matter of their constructions. Moreover, the term "ety-
mology" may be used loosely in such discussions, as though it too were a generic
designation. While it may not always be easy to distinguish etymology per se from
other forms of wordplay, I cannot agree with Barney (1998, 68) that it is "per-
haps unnecessary" to make the attempt. In my view, progress in the investiga-
tion of ancient treatments of naming would be facilitated by a closer examination
of the genus and species to which the evidence gives rise. The term "wordplay"
will suffice as a generic designation as long as one keeps in mind that such "play"
often issues from and reflects quite serious intentions. According to the param-
eters of the current inquiry, etymology, that is, the underscoring of derivations
on the level of semantic constitution —as when Hera is traced to aer— represents

65. With regard to limits on flux's prominence in the etymologies, it is useful to keep in mind
the observation of Heath (1888, 203) that it does not come to the fore until the discussion of ta
kala onomata, which begins at 4113. My concern here is with Plato's interpretation of Heraclitus'
stance. Consideration of the extent to which Plato's strong emphasis on the prevalence of flux in
describing Heraclitus' view reflects the latter's own position would take one far afield in the present
context.
66. See also Diels igro, 3.
 ETYMOLOGY AND THE CRATYLUS SOURCES 71

one prominent subdivision of this category. Other notable species include

eponymy, judgments of fitness involving functional terms, stress on dual mean-
ings of single terms, cases of synonymy, and the underscoring of parallels on the
phonetic plane. While Plato inherits a state of affairs in which distinctions
between these phenomena are not firmly in place, his own practice evinces a
strong interest in taking the discussion to a new level. Key support for this claim
is provided by the fact that Plato is unmistakably aware of sophistic construc-
tions oforthotes and at pains in the Cratylus to mark his approach off from them
at the species level. Moreover, although the Cratylus introduces etymology,
eponymy, and the use of functional terms, in its consideration of orthotes it ad-
heres firmly to the first of these —reserving for other occasions independent,
individual discussion and assessment of the latter two devices; Plato's procedure
indicates his keen interest in distinguishing notions that were not similarly dif-
ferentiated by others, in this case the literary tradition.
A key question regarding those Heraclitean fragments in which etymological
play has been suspected is whether the linguistic connections actually cross over
into that arena or are limited instead to the phonetic plane. A crucial reason for
uncertainty is that one typically cannot avail oneself here of the sorts of contex-
tual indicators that are so useful in the analysis of literature. 67 While one cannot
exclude the possibility that derivations are intended in certain fragments, such
attributions of source, if intended, are not obvious. Consideration of surviving
evidence does not rule out a Heraclitean concern with that type of derivation at
issue in the etymological species of wordplay, but neither does the material offer
confirmation of a marked interest in it. In what follows, I comment on DK 5,15,
25, 23, 48, 28,114, and 32.
According to DK 5, "They are purified in vain with blood, those polluted with
blood, as if someone who stepped in mud should try to wash himself with mud.
Anyone who noticed him doing this would think he was mad"

67. Regarding Heraclitus, cf. Dalimier 1998, 24-25. While in certain literary passages it is
difficult if not impossible to determine whether the level of play involved is semantic or pho-
netic, in most instances a decision may be reached. To a significant extent this is because litera-
ture regularly provides a range of cues that signal the presence of a derivation rather than the
mere existence of common ground on the phonetic plane. Such indications, offered either indi-
vidually or in various combinations, include the relevant use of language of naming, where original
assignments, name changes, or nicknames are involved (e.g., pseudonumos plus kalousin); terms
such as houneka, which signal explicitly that an author is identifying the source of an assignment;
and compression, in the sense that authors' remarks include the placement in close proximity to
one another of a name and two onomata whose consolidation yields that name. For examples of
passages that feature these cues, see Horn. Od. 19.407-9, 20.288-90; Hes. T/i. 144-45,195~9^> 207-
10, 281-83, °P- 8°-82; Pind - Of. 6.53-57,lsthm- 6-53; Aesch- Sept. 577-79 (with 658), PV 85-87,
732-34, Cho. 949-51; Soph. OT 1032-36, OC 486-87, A;. 430-32; Eur. Tro. 989-90, Ion 661-63
(with 800-802, 830-31), FT 32-33, Phoen. 636-37 (with 1493), Hel. 8-14, 1301-2 (with 1320, 1327
28, 1340, 1356); Hdt. 2.52.1, 4.107, 5.926, 6.63.3, 7.193.2, 9.91-92.
72 THE ANCIENT QUARREL

Here the term used to designate the state of mind of people who try to counteract
religious pollution by engaging in the same type of activity that generated the con-
dition is phonetically similar to that which denotes the condition they seek to rem-
edy. It is thus not simply because these people are polluted (miainomenoi) that
they would be judged as mad (mainesthai): Heraclitus' formulation, along with
the comparison involving mud, makes this clear. If the connection had been
direct, then one might have treated this fragment as similar to a passage in Pindar's
second Pythian ode (90-92), where the poet says that the envious "pull
on a line unwound to the limit until it snaps: a wound to their own heart
before they compass their desires." In Pindar, in contrast to Heraclitus, the basis
of connection is that of cause and effect since it is individuals' performing the
action indicated by the participle that results in their suffering the injury signi-
fied by the noun. While the presence of this connection might lead one to con-
sider whether a derivation of helkos was being suggested by Pindar, one has no
comparable impetus to speculate about a possible etymology in the case of
Heraclitus.69 I must therefore concur with the judgment of Marcovich (1967,
460) that "an intended word-play . . . is not likely" here. More specifically, fram-
ing the point in light of the genus-species framework introduced above, I would
suggest that there is a form of wordplay here, but that it is phonetic rather than
semantic in nature.
DK 15 runs as follows: "If it were not Dionysus for whom they march in pro-
cession and chant the hymn to the phallus, their action would be most shame-
less. But Hades and Dionysus are the same, him for whom they rave and cele-
brate Lenaia"

What keeps activity that includes an entity whose onoma is a

cognate of aidos from being quite shameful is the identity of the god in whose
celebration the activity is undertaken. As Craik (1993, 55) points out, a close
connection between aidos and aidoia was already assumed by Homer, who sub-
stitutes the former for the latter on two separate occasions (II. 2.262, 22.75).70
Heraclitus, unlike Homer, has no qualms about using the term aidoia. Indeed,
he is concerned to make the point that if viewed in the correct light, the activity
in question is not shameful. To all appearances, the province of Dionysus is life.
Yet in Heraclitus' view those practices that allegedly celebrate the god, promi-

68. Unless otherwise noted, translations of Heraclitus are those of Kahn (1979).
69. I am suggesting, not that Pindar offers a derivation, but only that his formulation makes
it more reasonable than in the case of Heraclitus for an interpreter to wonder, at least initially,
whether an analysis was intended.
70. According to Richardson (1993, 114), "Homeric decorum generally avoids references to
al8oia." Craik (1993, 55) observes that "in terms of linguistic usage, 018(65 may be viewed as the
natural reaction to the cdSota (and, of course, to people who are ai8otoi), just as yehoK, is to
yeA,oia."
 ETYMOLOGY AND THE CRATYLUS SOURCES 73

nent among them drunkenness and orgiastic conduct, are associated in crucial
ways with death.71 The intoxication prevalent during the Lenaia would dull
people's rational faculty and as such be a distinctly negative event.72 Intoxica-
tion is correlated with orgiastic conduct, which may result in generation. Although
reproduction might appear to be a purely life-affirming act, it is also correlated
with progenitors' diminution and ultimate demise (DKio). 73 From this perspec-
tive, it might seem as though people's behavior with respect to Dionysus was
irreverent: while the god represents life, their supposed acts of tribute evince a
strong tie to death. Yet once one apprehends the close cosmic interrelationship
between life and death — emphasized elsewhere (DK 88,62) and personified here
by the "identification" of Dionysus and Hades—people's conduct, far from being
irreverent, or shameful (anaidestata), becomes both intelligible and perfectly
appropriate.74 With Robinson (1987, 87), I do not see that the etymological tie
between Hades and aides is pertinent to the fragment's interpretation.
In DK 25, Heraclitus says that "greater deaths are allotted greater destinies"
His repetition of mu and mezo
is an obvious instance of play on the phonetic level. The point of contention is
the status of moms and moira. The phenomena are also related—without the
provision of an etymology—in Aeschylus' Libation Bearers, where Orestes tells
Clytemnestra that "destiny has so wrought that this shall be your death" (Tov8e
Heraclitus' treatment displays the revers
of that priority evident in Aeschylus: the tragedian makes fate dictate the mode
of one's passing, but Heraclitus, in keeping with his greater stress on individual
effort and responsibility (as, e.g., in DK 119), insists that one's destiny is shaped
by how one dies. With regard to the linguistic plane, he is surely cognizant of
the phonetic ties between moms and moira. Any interest in their relation
as cognates takes the form of an assumption rather than a discovery.75 While
the absence of context here makes conclusive statements difficult, what
Heraclitus says does not indicate that he is treating moms as the source-onoma
of moira.

71. Cf. Robinson 1987, 86 n. i: "Heraclitus is at pains to stress the frenzied and orgiastic (ie,
'Dionysiac') behaviour of the participants [in the Lenaia], not simply their drunkenness. But
drunkenness. . . there undoubtedly was, and the assumption of its existence—and consequences-
seems necessary for a complete understanding of Heraclitus' remarks."
72. In Heraclitus' view, "a man when drunk is led by a beardless boy, stumbling, not perceiv-
ing where he is going, having his soul moist" (DK 117). He maintains elsewhere that the dry soul
is "wisest and best" (DK 118) and that "for souls it is death to become water" (DK 36).
73. Cf. Kahn (1979, 264), who observes that according to Heraclitus "the desire of men 'to
live . . . and to leave children behind' is really a desire for their own death and replacement."
74. For emphasis on the point that here, as elsewhere, "the same" for Heraclitus does not
mean "identical," see Robinson (1987, 86), who notes that "at best it suggests a doctrine of'per-
spectival identity' (= interconnectedness of apparent opposites)."
75. Cf. the comment of Robinson (1987, 92) that the fragment "exploits the common etymo-
logical origin of the nouns moros . . . and moira."
74 THE ANCIENT QUARREL

DK 23, in turn, involves a conceptual link in the form of a contrast depen-

dency: "If it were not for these things, they would not have known the name of
76
Justice" One may accept here
the interpretation of Kahn (1979, 185), according to which "the conceptual
dependence of justice upon the existence of injustice and legal disputes . . . is
expressed not in terms of concepts but in terms of the name by which Justice is
known. If there were no judgments and penalties, men could not know or under-
stand the word dike that denotes them. But then they would not know the name
of Justice."
More relevant for current purposes is Heraclitus' famous observation (0X48)
that "the name of the bow is life, but its work is death"
the same tie is emphasized in Sophocles' Philoctetes (931-
33). Marcovich (1967,192) takes this fragment to illustrate the fact that accord-
ing to Heraclitus a name "reveals a great deal of the true of its object." I
would suggest, in contrast, that Heraclitus capitalizes here on the fact that the
words "life" and "bow" are identical in spelling to make a point
most fundamentally about opposition between language and reality, or the effect
the bow has on elements thereof. Rather than using this parallel as strong evi-
dence of a belief in particular names' revelatory power, one might conclude from
the fragment's content that a focus on linguistic connections obscures as much
as, or perhaps even more than, it clarifies (this point is emphasized by the pres-
ence of de). On the formulation of Snell (1926, 369), "the name conveys. . . the
opposite of what is essential." Even if one accepted a stronger tie here between
the bow life (Pioc; onthee strength of other fragments treating life
and death (DK 62, 88), there is no sign that Heraclitus wishes to derive the former
onoma from the latter in the manner at issue in Plato's Cratylus.
In DK 28, Heraclitus says that "the great man is eminent in imagining things,
and on this he hangs his reputation for knowing it all"
Despite his having an elevated reputation, which would
ostensibly correlate with a superiority over ordinary mortals regarding knowledge,
in Heraclitus' view such an individual is not justifiably invested with this dis-
tinction. As Kahn (1979, 211) points out, here "the eminence of the man who
enjoys public recognition and approval (ho dokimotatos) is contrasted with the
shabby credentials of what he himself recognizes and accepts: dokeonta, mere
seeming or imagining." It is, of course, irrelevant to one's judgment about the
possible existence of an etymology in this fragment that dokimos (superlative
dokimotatos) actually has dechomai as its root. What matters is only the views of
Heraclitus and those whose judgment he is impugning. IfHeraclitus is taking as
his backdrop a common view that dokimos has this root, then he could be sug-
gesting that dokimotatos instead has dokeo as its source and for this reason is a

76. For an emphasis on contrast dependency, see also DK 111.

77. Tr. Marcovich (1967).
 ETYMOLOGY AND THE CRATYLUS SOURCES 75

pejorative term, not a complimentary one. This derivation, if proffered, would

be of interest vis-a-vis the Cratylus, yet the text of the fragment, while not ruling
out this interpretation, also does not provide clear confirmation of it.
According to DK 114, speaking with intelligence necessitates the apprehen-
sion of what is common (i.e., the logos). Here the phonetic similarities in
Heraclitus' twofold use of the dative (xun nooi and xunoi), along with the com-
bined presence of ischurizesthai and ischuroteros, reinforce the substantial claim
being made in the remark that contains it: "If people are to speak with under-
standing, they must hold fast to what is shared by all, as a city holds to its law,
and even more firmly"
.78 According to the etymologi-
cal paradigm in literature and the Cratylus, xunoi would have to represent a
consolidation of xun nooi. Yet the causal dependency here points in the other
direction, insofar as it is individuals' proper mental focus on the logos that enables
their speech to embody understanding. Thus, while this fragment, unlike DK 5,
involves a causal relationship between referents of phonetically similar words,
the very presence of this tie works against the idea that Heraclitus' statement
involves etymology. Since I find no contextual cue suggesting that he offers an
etymology, and a salient indication that the attribution of such a derivation would
be implausible, I must agree with Kirk (Kirk, Raven, and Schofield 1983,210 n. 2),
who notes that while xun nooi-xunoi is an instance of wordplay, it does not con-
stitute an example of "etymological periphrase."
It is worth mentioning, finally, DK 32: "The wise is one alone, unwilling and
willing to be spoken of by the name of Zeus"
As Marcovich (1967, 445-46) and Kirk
(1962, 392) have noted, the presence of an etymological tie to zen is unlikely.
Although this link, which has a strong presence in literature, cannot be ruled
out, it would be a stretch to tie hen to sophon closely to the notions of life and
death in the way required by this interpretation.
While literary authors resort frequently to etymology, and make substantial
use of it to treat questions of names' appropriateness, study of the extant frag-
ments does not offer obvious support to the claim that Heraclitus exhibits a
marked interest in this particular device. In one's making of judgments about
the presence or absence of derivations, context is all-important. As long as this
context is lacking, inferences about what is missing in a given fragment must be
made by interpreters (see, e.g., Kahn [1979, 185] on adikia in DK 23), perhaps
erroneously. Although certain fragments are intriguing from the Cratylus' per-
spective, clear determinations of the level of wordplay are difficult, if not impos-
sible, to make. Quite apart from this, militating strongly against Heraclitus as
the inspiration for Plato's discussion oforthotes onomaton is individual natures'

78. Tr. Kahn (1979), modified to underscore the conditional force of the claim about
understanding.
76 THE ANCIENT QUARREL

prominence in the Cratylus as what etymologies reveal.79 I thus do not dispute

the view that Heraclitus is a central Platonic opponent in the Cratylus but would
maintain that, based on surviving evidence, he is demonstrably such only on
the ontological plane, that is to say, with respect to the doctrine of flux that comes
to prominence after the discussion of etymology is well under way.
I turn lastly to the Derveni papyrus, which was discovered in northern Greece
in 1962.80 According to Laks and Most (1997, 5), its author is the "spokesman"
for a "sophisticated version of Orphism," in which "religious initiation takes the
form of a rationalizing exegesis of the episodes of an Orphic theogony set into
systematic relation with the episodes of a late Tresocratic' kind of physical cos-
mology."81 The text contained in the papyrus was composed around 400 B.C.,
and the Derveni copy had its genesis "some decades" thereafter (West 1983,82).82
The Derveni papyrus reflects the impact of several Presocratics, including
Heraclitus, Anaxagoras, and Diogenes of Apollonia.83 Yet, as Burkert notes, it
exhibits no Platonic influences (1986, 4; 1968, 100), and it is unclear whether
Plato was acquainted with it (1997,174). The relevance of the papyrus has been
emphasized by Baxter (1992), who argues that it "remains a prime candidate as
a target of the Cratylus" (139).
The mere presence of a concern with linguistic issues is not in itself a basis
for asserting anything more than a general link of the sort that is also present in
other cases. As we saw with regard to the sophists, the key question is what rela-
tionship emerges once one moves to the level of specifics.84 In elaborating his

79. Cf. Snell's observation that, far from endorsing the view that there are individual divine
natures, Heraclitus "seeks the all-encompassing 'divine' and construes the divine as a unity" (1926,
374)-
80. For discussion of the circumstances, see West 1983, 75-77; Funghi 1997, 25.
81. For an investigation of the role of allegory in the papyrus, see Laks 1997. Commentators
diverge on the issue of whether Euthyphro might be the author of the papyrus. According to Kahn
(1973, 156 n. 6), "It seems natural to suppose that Euthyphro . . . was known to be given to alle-
gorical explanations of divine names in the style of the Derveni papyrus"; indeed, he "might even
be its author" (see also Kahn rg86, 98-99, and 1997; Rosenstock 1992, 404 n. r8; Funghi 1997, 36;
Tsantsanoglou 1997, 122 n. 54). In contrast, although Baxter (1992) believes that the Derveni pa-
pyrus "remains a prime candidate as a target of the Cratylus" (139), he regards the possibility that
Euthyphro might be the writer in question as "far-fetched" (133).
82. On the date of ca. 400 for the text itself, see Henry 1986,150; Burkert rg68, roo, and 1970,
443; West 1983, 77-82. Burkert (1983, 38) gives the approximate date as 400/380. In a fourth article
a few years later, where Burkert argues that Stesimbrotus is the author, he suggests a range of 420-
370 (1986, 4).
83. For discussion, see Burkert 1968, 1983.
84. As of this writing, there is no complete published edition of the papyrus. The following
remarks about its content therefore have as their foundation the translation of Laks and Most (1997),
the first version of which was checked by K. Tsantsanoglou, who also offered suggestions based
on his supplements to the extant text (Laks and Most 1997, 9). My provisions of Greek beyond
what is included in Laks and Most (10-22) are based on the text that appears in Zeitschrift filr
Papyrologie und Epigraphik 47 (1982) following p. 300. The translations and column numbers
provided in the following remarks are those of Laks and Most.
 ETYMOLOGY AND THE CRATYLUS SOURCES 77

own particular brand of Orphism, the Derveni author evinces a marked interest
in the genesis and operation of the cosmos and in connections between language
and reality. Most striking is the emphasis on the absence of distinctions where
ordinary people regularly assume them to exist (for the underscoring of a gap
between common and superior understanding, see col. XVIII; cf. col. XVII): on
the formulation of Edwards (1991,210), this sophisticated Orphic "will never see
distinction where distinction can be avoided." Underscoring the same point with
explicit reference to the divine, Tsantsanoglou (1997, 99) observes that the author
of the papyrus "does not deny the existence of god, in the singular, whom he
mentions in the second part of his book, but gods in the plural are referred to only
in relation with other people's beliefs or practices." As Funghi (1997, 35) indicates,
this god or divine principle that produces and governs the cosmos "is identified
with Zeus, and other divinities or abstractions are denied except as his attributes."85
Such formulations make it clear straightaway that the ontological foundation of
the Derveni author's remarks is quite different from that view, pivotal to the
Cratylus, according to which divine onomata have distinct referents and there-
fore natures, which are disclosed via analyses of their descriptive content.
While the unenlightened assume without question that each divine onoma
has a distinct referent, the author of the papyrus insists that names such as "Ura-
nus," "Cronus," and "Zeus" in fact pick out a single entity. 86 Moreover, in
column XXIII he contends that "Ocean is the air and air is Zeus," and on a basis
that is not made entirely clear (col. XVIII; cf. West 1997, 87), Moira and Zeus
are the same.87 In column XXI, in turn, "Heavenly Aphrodite and Zeus (and to
aphrodise and to jump) and Persuasion and Harmony are established as name
for the same god " Elsewhere (col. XXII), the
Derveni author states that Earth (Ge or Gala), Mother (Meter), Rhea, Hera,
Demeter, Deio, and Hestia are names for a single divine entity; Burkert (1986, 3)
adds that "in this collection Leto too would fit easily." Finally, as Edwards (1991,
210) observes, the author contends (col. XXVI) that "such a locution as ^[ipjoq
eaq cannot signify 'his mother' (and thus imply a separation of masculine and
feminine) but 'Mother of good things.' . . . The mothering god, no less than Zeus,
is Mind."
In addition, the papyrus stresses equivalence of meaning in the case of verbs
thought, again by the unenlightened, to differ in meaning. Its author's empha-
sis here is sharply opposed to that of Prodicus, who foregrounds differences of
meaning between terms regularly thought to be interchangeable; while the two
agree in distinguishing ordinary from sophisticated usage of a range of terms,

85. Cf. Merkelbach 1967,21: "In the commentary, a doctrine is represented according to which
'Zeus' is the principle of all existing things. All gods are to be traced back to Zeus; all elements,
and even their combining—which is called poetically 'Aphrodite'—are the same Zens."
86. Col. XV states the identity of Cronus and Zeus. On Uranus' identity with Cronus, see
Rusten's discussion of col. XIV (1985, 134-36).
87. On air and Zeus, see also cols. XVII and XIX.
/8 THE ANCIENT QUARREL

they thus differ substantially in their conceptions of what is most characteristic

of deficient practice. In column X, for example, the Derveni author first equates
"saying" and "uttering" ), then "say-
ing" and "teaching" . 88 Similarly,
in column IX, as Rusten (1985, 128-29) observes, the text offers another three-
fold equivalence, namely, lambanein = haptein = epikratein. Elsewhere (col. XI),
the author of the papyrus contends that "'proclaiming the oracle' and 'prevent-
ing harm' have the same meaning (jcpfjccii Se KOQ, dpKeoai Tamo [8i>]vaTai)"
and equates distrust with ignorance (col. V). In column XXI, in turn, he claims
that Peitho = peithein since eikein = peithein (cf. Henry 1986,157). I cannot agree
with Baxter (1992, 133-34), who underscores what he views as the similarity
between the text's provision of multiple onomata for the same referent and Plato's
approach in the Cratylus. In my view, Baxter overemphasizes the parallel inso-
far as Plato considers instances in which the sorts of ontological distinctions
denied in the papyrus (e.g., between deities) are assumed and stresses that the
same nature can be described by different sets of linguistic components ("Hec-
tor," "Astyanax," and "Archepolis," e.g., are all appropriate names for one whose
societal role is that of protector, 39/|.b-c).
Where allegory is in evidence, etymology may, though need not, be found as
well. While the latter does make an appearance, Kahn (1997, 61-62) notes rightly
that etymologies in the Cratylus' sense are not a major presence in the papyrus:
"The Derveni author does not in general practise etymology in the Cratylus
manner, reconstructing the 'original' and more instructive form of a familiar
word." The two cases in which direct comparisons are possible are, in his view,
those of Cronus (col. XIV) and Demeter (col. XXII).
According to Kahn (1997,62), while the Derveni analysis of the name "Demeter"
("Mother-Earth") is "banal," Plato's "seems by contrast deliberately far-fetched."89
Baxter tentatively associates the Cratylus' derivation with the Derveni papyrus
specifically or Orphic "scholars" more generally (1992,132), although it can be
tied clearly to Euripides.90 Plato's analysis finds parallels both in literature and
in the papyrus, but in its details it is closest to Euripides, in particular his rela-
tively complex depiction in Helen, and may be viewed as simply a more explicit
formulation of the idea at issue there. A connection exists between the Cratylus
and the papyrus in the case of Cronus, insofar as both derivations include the
term nous. On a more specific level, though, the two analyses diverge: whereas
the Derveni author treats the divine principle at the cosmic stage in question
as the Mind (nous) that strikes (krouon) things against one another, on Plato's

88. See Rusten (1985, 131) for a reconstruction of the author's reasoning.
89. Although I concur that Plato's derivation is more elaborate, I cannot agree with the con-
tention of Henry (rg86, rj8) that the Cratylus' explanation of the etymology is different.
90. Less plausibly, given the presence of tighter connections in other cases, Goldschmidt (1940,
125) links the Cratylus' derivation to Prodicus' famous remark (DK 5) that human beings tend to
deify that which confers benefits upon them.
 ETYMOLOGY AND THE CRATYLUS SOURCES 79 79

analysis the name signifies "the pure and unmixed character of the mind itself
(TO KaGapov amov Kai rJKipaTOv wv vov, Cra. 396b6-y). One may note, in
addition, the Derveni author's statement, following his etymology of "Demeter"
(col. XXII), that "she is called Deio too, because she was cut (e5r|u66r|) during
sexual intercourse."91 In column XXI, moreover, the author observes that the
cosmic principle was called "Harmony, because . . . she fitted together (-r|p|j,ooe)
. . . to each of the things that are." Even Baxter (1992), who generally seeks to
underscore the papyrus's connection to the Cratylus' inquiry, grants that in this
case one may speak only in a qualified sense of etymology (132 n. iry). 92
Although etymologies are not a major presence in the Derveni papyrus, one
may tie it more closely to the Cratylus in this regard than to any other source
discussed here apart from literature. Yet the foundation of Plato's inquiry diverges
sharply from that of the papyrus. Its author distinguishes the superior understand-
ing possessed by himself from that level of apprehension typical of most human
beings. While they generally assume that each onoma — above all, one that is
divine—has a separate referent, one with knowledge is aware that this is decid-
edly not the case: according to the Derveni author, far from being fundamental,
individual divine natures do not even exist. From his perspective, literary authors
would be pivotal, and highly influential, examples of those operating in the
absence of understanding. Far from considering whether onomata like "Demeter"
and "Zeus" have the same referent, poets assume distinctness and explore the
individual natures of these and other deities via the descriptive content of their
names.93 Burkert (1970, 450) compares the Derveni author's linkage of cosmic
genesis and naming with that of Hesiod ("In Hesiod . . . one finds in some situa-
tions new significant names"). More salient than the parallel, I submit, is the
crucial difference in underlying conceptions of reality and of what names may
be taken to disclose. Most importantly, it is precisely the literary perspective —
according to which onomata have distinct referents, whose natures may be dis-
closed via etymology—that undergirds Plato's own inquiry in the Cratylus.

91. Since, as Tsantsanoglou (1997, 99) points out, "gods in the plural are referred to only in
relation with other people's beliefs or practices," this statement cannot reflect the Derveni author's
own allegiance to a view that would stipulate the involvement of multiple divinities.
92. The Derveni author's handling of Aphrodite and Peitho differs from that of Harmonia. In
the former cases, the author's treatment of names moves through synonymous pairs of verbs, each
of which contains one member that is obviously equivalent to the name in question. Thus, in the
case of Aphrodite —i.e., Common Aphrodite, a la Symp. i8od-e, since Heavenly Aphrodite has a
distinct standing in col. XXI—meignunai is the intermediary and equated with aphrodisiazein. Simi-
larly, the divine principle was called "Persuasion, because the things that are yielded to one another;
and 'yield' and 'persuade' are the same"; that is, as previously noted, Peitho = peithein because eikein
= peithein. The author's assumption is that the proper name and verb in each case arc interchange-
able—with the former representing a personified version of the latter—and the author clarifies the
basis of this equivalence through intermediate terms unrelated in descriptive content to the others.
93. In a related vein, the Derveni author "reproves the vulgar notion of succession among
the gods" (Edwards 1991, 211), which plays such a pivotal role in Hesiod's Theogony.
 3
THE STRUCTURE AND OUTCOME
OF THE CRATYLUS' INQUIRY

i. Introduction
As we have seen, the literary tradition makes extensive use of etymology to arti-
culate the semantic constitution ofonomata, largely proper names, with the goal
of shedding light on bearers' natures —the existence of which is assumed.1 More-
over, authors make numerous judgments of the appropriateness ofonomata to
their referents. At the outset of the Cratylus, Plato "invests" naming qua etymolo-
gizing with techne status, providing a laudatory description of the sort one would
expect based on those criteria put forth in the Gorgias to distinguish genuine
technai from empeiriai. As the dialogue unfolds, however, Plato rejects the lofty
claims he had Socrates make for this activity and its practitioners. As in the
Gorgias, where Plato's focus is on rhetoric, here too what is identified at the outset
as a techne later has that distinctive status challenged decisively. In the Cratylus,
the literary tradition provides Plato with key material used in formulating this
challenge; in the course of his investigation, however, Plato rejects that tradition's
procedure, along with the chief assumption on which its use rests. Far from being
limited in its pertinence to the dialogue's early analogy between shuttle construc-
tion and naming, the techne issue, construed in the manner indicated above,
governs much of the discussion and is crucial to its unity.
Plato's critique of naming in the Cratylus is a central component of a larger
project. With a view toward defending philosophy's claim to govern content and

i. As noted in chapter i, this tradition does not have a technical philosophical notion of indi-
vidual natures; instead, authors work with a loose, nontechnical notion according to which a mortal
or divine individual is widely recognized by a salient characteristic, power, or kind of activity.
Individual natures, thus construed, play a central role in the Cratylus; for the phuseis of mortal
or divine individuals identified as the objects of etymological analysis, see, e.g., 394e8-rr, 39532-
3, 1)2—8, d3~5, 39632—bi.

80
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS' INQUIRY 8l

priorities with respect to values, Plato must criticize poetry and sophistry or rheto-
ric, which promote conceptions of flourishing (eudaimonia) that he finds to be
untenable. In the Ion, Protagoras, Gorgias, and Republic, the concept of techne
plays a crucial role in Plato's attempts to show that they do not merit the esteem
in which they are presently held and that, contrary to popular belief, philosophy
is the most laudable practice of all. Phrased in terms of the concept of techne,
Plato's goal is to show that the former are not technai, while the latter, in con-
trast, is the techne par excellence. His demonstration in the Cratylus that nam-
ing as practiced by the literary tradition is not a techne is integral to the project
of showing that poetry itself, in which issues of naming figure prominently, does
not have techne status. In this chapter I discuss the aspect of Plato's critique that
involves the Cratylus. In chapter 5 I explore the Republic's comprehensive chal-
lenge and its complementary relation to the Cratylus' discussion.
Opinions about the Cratylus' content and philosophical import are, like those
about its sources, quite divergent. Some interpreters have questioned the
dialogue's seriousness of intent and judged it to be of minimal philosophic sig-
nificance. Taylor (1960), for example, labels the Cratylus a "minor Socratic dia-
logue" (75) that offers "a picture of Socrates in one of his more whimsical moods"
(78). According to Fowler (1926,4), it "cannot be said to be of great importance
in the development of the Platonic system, as it treats of a special subject some-
what apart from general philosophic theory." Others, in contrast, focusing on
the opening section of the Cratylus (383-390) and on Plato's treatment of ety-
mology or phonetic constitution (i.e., prota onomata), conclude that the dia-
logue makes serious philosophical proposals.2 Such commentators rightly empha-
size the seriousness of Plato's intentions in the Cratylus. As I will argue, however,
these intentions do not result in the elaboration there of substantive positive
proposals. Instead, following a protracted investigation, Plato rejects approaches
to the notions of naturalness and appropriateness that are based on the constitu-
ency of onomata. Although the dialogue's emphasis is critical, Plato provides
indications there of the direction that his positive theorizing will take. A discus-
sion of those clues supplied in the final portion of the dialogue will offer a point
of departure for the consideration of Plato's favored approaches to orthotes
onomaton in the following chapter.

2. In using the phrase "phonetic constitution," I follow Schofield 1982. For interpretations privi-
leging a supposed positive stance, which rely heavily on Plato's remarks in the first section of the
dialogue, see Kretzmann 1971, 1967; Silverman 1992; Weingartner 1973; Kato 1986; Grote 1867.
According to Grote (1867, 525), Plato advances the etymologies "as genuine illustrations of a theory
of his own respecting names." Lorenz and Mittelstrass (1967) maintain that Plato is making "a pro-
posal for a rational reconstruction of language" (13) and that "the quasi-etymologies in the Cratylus
. . . are used as examples of how one should proceed when this programme is accepted" (10). Steinthal
([1890] 1961,99,108), in turn, suggests that Plato would have liked to institute a science (Wissenschaft)
of etymology but felt incapable of so doing. For accounts that focus instead on Plato's discussion of
phonetic constitution, see Jowett 1892, vol. i, 259-62; Leky [1919] 1967, 54.
82 THE ANCIENT QUARREL

2. The Techne Issue in the Cratylus

Though commentators have remarked on Plato's use of a "craft analogy" in fram-
ing the Cratylus discussion of naming, they have not recognized that his early
remarks purport to establish a high rank for naming, that of techne, which will
be systematically undermined in what follows; even those who include the term
techne in their remarks have not identified the specific form assumed by Plato's
concern, which links the Cratylus closely to the Gorgias} As I argue in what
follows, the techne issue, thus construed, governs a substantial portion of the
Cratylus and makes a pivotal contribution to its unify.
In Plato's dialogues, the term techne is employed in broad, or nontechnical,
and narrow senses. Instances of the broad meaning reflect ordinary usage, according
to which the term applies to a heterogeneous group of activities thought to involve
some type of expertise (i.e., in this context, use of the word is not based on a
systematic inquiry into the nature of its referent).4 When speaking more strictly,
Plato employs a distinction between technai and empeiriai in order to differen-
tiate what he takes to be quite disparate sets of activities.5 Plato grants that in
some sense both types of activities involve skill, but central to his project is care-
ful attention to the issue of what being a genuine expert entails. When deployed
for philosophical ends, the notion of techne thus becomes highly normative, and
the appellation one that is properly applied to only a select group of practices.6
Although central techne requirements are introduced quite early,7 it is in the

3. Thus, in the view of Kato (1986, 59-63, 69), for example, Plato endorses the view that
naming, or language generally, is a techne. Baxter (1992) broaches the subject of techne (45,
51), but his interpretation differs significantly from my own; notably, rather than Plato's discus-
sion of phonetic constitution applying a methodology previously set forth (77), I would main-
tain that this section, like the one centering on etymology, treats a constitution-based approach
to correctness that Plato finds untenable. Pfeiffer (1968, 57-65) comments on epistemai and
technai in his discussion of the Cratylus, yet does not identify the aforementioned dynamic at
work therein.
4. The heterogeneity of ordinary usage is strikingly evident in a famous Aeschylean passage,
where the god Prometheus is treated as the source of all human technai (PV 442-506). For Pla-
tonic examples of nontechnical usage, see Symp. 18764-5, where cookery, dubbed an empeiria at
Grg. 462d9-ri, is identified as a techne; with regard to tragic poetry, whose techne status is denied
at Grg. 502b-d, see Symp. 22-^Ay, though Grg. 5oier-7 denies that harp-playing and flute-playing
are technai, they are affirmed to be such at Euthd. 28901-4 and Prt. 32338-9, respectively; soph-
istry, grouped with empeiriai at Grg. 463^-6, is called a techne at Prt. 3^)3-4 and 31701-2; and
prior to its explicit classification as an empeiria at Grg. 46zb-c, rhetoric is identified repeatedly
as a techne (at 44809 it is characterized by Polus as "the finest techne of all").
5. For the Gorgias' distinction between empeiria and techne as originating with Plato, see
Capelle 1922, 262—65; Heinimann 1976, 144—45 (with n. 50). Cf. Allen (1994, 83), who notes that
"the sharp line between experience and knowledge drawn [in the Gorgias] was an innovation."
6. The term techne is often translated "art" or "craft"; more accurate for Plato in this context
would be something like "rational purposive discipline."
7. In the Ion, it is because rhapsodes and poets do not operate with understanding, and their
disciplines lack genuine subject matters, that they are not viewed as practitioners of technai. For
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS' INQUIRY 83

Gorgias that Plato provides a systematic account of the criteria that distinguish
genuine technai from pseudo-technai, or empeiriai. He presents there five condi-
tions, of which the three most salient are as follows:

1. Peri ti, or "aboutness": The activity in question requires a genuine "subject

matter." This subject matter includes both the Form(s) in question and a speci-
fication of that toward which activity is directed.8
2. Understanding: Those engaged in the pursuit must be in the proper cognitive
state, that is, operate with episteme ("knowledge" or "understanding") rather
than mere belief;9 presupposed here is the existence of a small class of expert
practitioners. One with understanding of a given subject matter will be able
to provide a rational account (logon didonai) of those procedures that com-
prise the activity and their relevance to the end in question.
3. Goodness: As Moravcsik (1992, 15) observes, "the activity or its objects —or
both —must be good or lead to what is good." Where human beings are
involved, the idea is that the practice in question must be of genuine benefit
to them, not simply cater to their whims and indiscriminate desires.10

an emphasis on the understanding requirement, see 53/|.b-d, 535b-d, 5423. At 5356, it becomes
clear that Ion, though not emphron, is nevertheless quite capable of calculation regarding the
effect he produces on the audience. The idea that one who has good means-ends reasoning
skills may well lack genuine understanding is explored at far greater length in the Republic.
Plato combines discussion of the two conditions at 53yd-e, where he says that technai are
distinguished from one another based on the fact that their practitioners have episteme of dif-
ferent subject matters (cf. 5383). He stresses the importance of the peri ti condition at the
dialogue's end when Socrates says that, far from giving him a display of his episteme regarding
Homer, Ion is unwilling to specify "which things you are clever about" (54165-6). For an em-
phasis on this dialogue's importance to the understanding of Plato's conception of techne, see
Kahn 1993.
8. While the Gorgias does not specify the ontological status of the entities in question, later
dialogues make clear that Forms play this role. On the Gorgias' stance, see Moravcsik 1992, 2
(cf. 24): "Plato does not elaborate the criterion for genuineness in this dialogue, but from the treat
ment of rhetoric and the way it contrasts with mathematics and medicine, one gathers that genu-
ine objects must have permanent and stable natures and must embody such characteristics as
order and harmony."
9. At Grg. 454c-e, Plato draws a sharp contrast between episteme and pistis. As Dodds (1959,
206) has noted, the knowledge-belief distinction is first made formally by Plato in this exchange
between Socrates and Gorgias. Subsequently, doxa becomes the standard term for opinion, and
in the Divided Line (Rep. 5ogd-5iie), pistis and eikasia are subdivisions of it. Plato considers
what understanding involves, and how it is related to belief, at far greater length elsewhere; with
respect to the distinction between technai and empeiriai, one may single out his systematic dis-
tinction between faculties and their proper objects at Rep. 475d~48oa.
10. Woodruff (1990, 73) notes that while the Gorgias makes explicit the principle that a techne
in the strict sense aims at the good, this is implied in earlier dialogues (he cites in this connec-
tion La. i85c-d and Euphr. 146). A fourth techne condition involves teachability (and, by impli
cation, learnability); in this formulation of the criterion, I follow Moravcsik (1992, 15). Finally,
there is a fifth and less prominent requirement that the activity in question represent some ca-
pacity (dunamis).
84 THE ANCIENT QUARREL

In the Gorgias and Cratylus, Plato first appears to assume the techne status of rheto-
ric and naming, respectively, then undermines their claims to that status as the
inquiries proceed. In both dialogues, the point of contention is whether the prac-
tice under investigation qualifies as a techne in the narrow sense of that term.11
At the outset of the Cratylus, Plato "asserts" that naming is a techne (3886). The
term to onomazein is ambiguous in a way that to kerkizein (weaving) and to trupan
(drilling) are not, since "naming," as such, encompasses both the construction of
onomata and their use. According to the particular theory of name use investi-
gated in the Cratylus, one decomposes onomata to determine what they describe;
that is, one engages in "name decoding." This theory connects name use to name
construction insofar as the criteria for names' proper employment must be part of
the telos in the "artisan's" mind. In considering whether naming qualifies as a
techne, Plato has in view construction conjoined with use as specified here.
That naming is alleged to be a highly normative enterprise is revealed not only
by Plato's identification of it as a techne but also, more specifically, by his early
invocation of the nomothetes as onoma constructor, along with his presentation of
the dialektikos as employing his knowledge of ousiai to determine whether the
resulting products are properly constituted. Judgments of appropriateness are based
on whether analyses of words' deep structure succeed in revealing the natures of
their referents; in this context, as for the literary tradition, "correctness in assign-
ment" (opGcoq e%ew) is equivalent to "being in accord with [a referent's] nature"
(mid ((wow eivoa) (3946-3953). Since, according to the natural-correctness the-
sis, onomata are fit for use to the extent that there is a match between their descrip-
tive content and the natures of their referents, and since the dialectician is the one
who evaluates what is produced (39002-11), one may conclude that the dialecti-
cian determines whether onomata are usable by judging in each case whether the
aforementioned correlation obtains. Plato supplies here the expected cognitive
underpinning for an activity of this caliber, vigorously denying any connection with
mere belief (doxa). At this early juncture, Socrates concludes that if these and
related contentions about the elevated status of the activity were well founded, it
would hold an impressive rank indeed, qualifying as a techne in the narrow sense
of that term. In what follows, however, Plato dismantles every one of the exalted
claims he has "made" for the practice of naming; hence, as in the Gorgias, what
purports initially to be a techne later has that status called into question.12

n. The view of the Gorgias assumed in this study, which emphasizes its ties to later dialogues,
in this case the Cratylus and Republic, thus differs from that of Hall (1971,202), according to whom
the Gorgias' closest ties are "to the utilitarian ethic grounded on the knowledge and values of
techne so characteristic of the early dialogues."
12. The hypothetical character of the discussion goes unremarked by those who treat what is
said early on as representing Plato's own stance. In the view of Kretzmann (1971), 3god-e is a
summary of "the allegorical presentation of Plato's general theory" (130), which according to Plato
is "an established truth and not a mere hypothesis" (133, italics mine). For other examples, see
Anagnostopoulos 1973, 329-30; Kahn 1973, 166 n. 18.
 THE STRUCTURE AND OUTCOME OF THE CRATYLUs' INQUIRY 85

Plato evaluates naming's claim to techne status in the course of his assessment
of the natural-correctness thesis. Before Plato can investigate the techne issue,
he must therefore indicate what he means by orthotes. In order to do this, as we
saw, Plato recurs to literary methodology and its assumption about the primacy
of individual natures (391^-3973). Following the specification of how he con-
strues orthotes, which completes the foundation for his inquiry into naming's
techne status, in the course of the investigation Plato finds cause to reject this
assumption and procedure. 1 '
As noted, a central condition on technai is that practitioners operate with
knowledge or understanding rather than mere belief. Since fulfillment of the
"understanding requirement" presupposes the existence of expert practitioners,
a single techne condition is involved. Also crucial is the peri ti, or "subject-
matter," condition. These two criteria are most prominent in the Cratylus, but
as the earlier discussion of etymology based on the Republic suggested, the good-
ness criterion too plays an important role in Plato's inquiry. Further treatment
of this requirement will, however, be reserved for chapter 5, which gives an
important place to the relationship between the Cratylus and Republic. For the
purpose of illustrating how the Cratylus' challenge to naming's techne status
unfolds, I concentrate in what follows on the understanding and subject-matter
conditions.
In the spirit of literary practice, the nomothetes, connected here and else-
where with the normative enterprise of giving laws, becomes also a word-giver
(onomathetes) based on a linguistic ploy (3886-3893).14 In the course of Plato's
investigation of naming's techne status, the nomothetes is shown to be a mere
construct that was introduced of necessity in the dialogue's early description of
word formation as a normative endeavor. Having initially depicted a single indi-
vidual as responsible for production, in what follows Plato vacillates in his state-
ments concerning just whom he envisions as having played that role, and with
what degree of success. Thus, Plato depicts the constructors ofonomata as "astro-
nomers and idle talkers" (iietecopoXoyoi twe<; Kai cc8oXea%ca, 40^8-9) and
likens them pejoratively to many of the current "so-called wise men" (4iib6).
Elsewhere, he attributes the invention of one term (soma) to the Orphic poets
(4OOC4-5).15 What is most significant is not, as Demand (1975,108) has claimed,
that Plato "often departs from the notion of a single Nomothetes to speak of a

13. While the literary tradition operates unsystematically and without conscious reflection
on its methodology, Plato draws on its assumption and procedure quite consciously, with the
ultimate goal of discrediting them.
14. This familiar role for the nomothetes is, naturally, prevalent in the Laws. See also Rep.
42734, 46234, 53005, 538d8, 56401, 59962; Pol. 29469, 305(35, C2, jogdi.
15. Concerning the Pythagorean and Orphic backdrop for the Cratylus' ideas about soma,
see Ferwerda (1985, 279), who argues that Plato "deliberately exaggerates the esoteric aspect of
their doctrines by implying that the Pythagoreans consider the body as the tomb (and not the
sign) of the soul and that the Orphics say that the soul is in the body as in a prison."
86 THE ANCIENT QUARREL

plurality of namegivers" but rather that, following his initial characterization

(3886-3906), Plato's valuation of those with this function shifts dramatically for
the worse, and at key points he speaks of them as belonging to different classes
of individuals. This is the crucial move since it involves the dissolution of the
initial link between the promulgator of norms and constructor ofonomata.16
Despite the pivotal role played by the dialectician in the enterprise of nam-
ing as initially described, the term dialektikos is used only three times in the
Cratylus; of these, only the two early instances are pertinent. 17 Providing a stan-
dard characterization, Socrates asks Hermogenes whether the dialectician is the
one with expertise in question and answer (390010-11; cf. Rep. 534d-e). In a
passage of greater importance for current purposes, the dialectician is portrayed
as the one directing and supervising the activity of the nomothetes (39od5-y).
The authority of tec hue practitioners—and of the dialectician most of all—derives
from their grasp of ousiai. Guided by this awareness, the dialectician's activity is
norm governed in the strictest possible sense; hence, it seems appropriate for Plato
to invoke this figure at the outset, when the haphazard character of the naming
enterprise has yet to be revealed. Despite the great fanfare with which the dialec-
tician is introduced, however, Plato quite abruptly ceases talking about him. If the
dialectician's activity were in fact central in the way initially suggested, Plato would
be expected, even compelled, to provide him with a more sustained presence in
the dialogue. A genuine techne, as characterized in the Gorgias and elsewhere,
cannot have random practitioners—on a cognitive and motivational par with rheto-
ricians, who are criticized in the Gorgias—but only those meeting very stringent
requirements. Plato's treatment of the legislator, supplemented by that of the dia-
lectician, shows that naming qua etymologizing has no practitioners of the requi-
site caliber and hence plays an important role in his challenge to its techne status.
As might be expected —since meeting the understanding requirement pre-
supposes the existence of a class of experts, which dissolves in the course of the
inquiry—the cognitive ground of the naming process shifts quite dramatically.
At the outset of the Cratylus, Plato repudiates the idea that doxa directs the enter-
prise of constructing and evaluating onomata (3873). On several occasions, Plato
utilizes the verb epistamai (see, e.g., 38gd, 3936). Moreover, he emphasizes
Socrates' sophia at diverse points in the inquiry (as, e.g., at 3960, 4016, 4106).
While Plato continually insists that activities are to be performed in the way
naturally appropriate to them and not based merely on individuals' own beliefs,
it becomes increasingly clear that it is actually belief rather than knowledge that
governs the practice of naming at issue in the Cratylus.

16. On the fate of the nomothetes, cf. Derbolav 1972, 63: "Once the insight is reached that
coming up with the correct word is not mainly an achievement of construction but is rather above
all a matter of having the right conception of reality, word formation is deprived of its emphatic
character and the Nomothetes loses the aura of a lawgiver."
ry. The third (at 3g8d6-y) occurs in a facetious reference to heroes as dialecticians.
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS INQUIRY 87

In fact, Plato's remarks on the governance of doxa become progressively more

general in scope. First, he states that insofar as Homer's views are under consid-
eration, what has been grasped is the poet's belief (doxa) "about the correctness
of words" (393b3~4); at issue here is the doxa of a single individual. Subsequently,
Plato observes that by analyzing divine names one gains awareness only of the
human doxa that produced those onomata (40134-6). Rather than simply gov-
erning one individual's approach to naming, here belief is identified as the source
of an entire class of assignments, one, moreover, of particular importance to the
literary tradition.18 Still later, Plato concludes that those first assigning all classes
of onomata did so based on a particular belief (doxa) they held about the char-
acter of reality (4iib-c). Notably, this comment introduces his discussion of "fine
names" (ta kala onomata), which includes numerous terms whose referents are
of central philosophical importance. Since, in Plato's view, doxa cannot grant
one access to reality, it is surely not coincidental that his final —and global —
generalization about the cognitive basis of earlier attempts at naming occurs at
this juncture: if belief lacks this capability, then surely no product to which its
employment gives rise will be able to yield understanding.
As previously noted, one with understanding of a given subject matter must
be able to offer a logos of those procedures comprising the activity and their rele-
vance to the end in question. In the case of naming, the key questions are: what
methodology is followed in the construction of words, and how does it promote
the goal of creating onomata that disclose natures? In the Cratylus, the genesis
of onomata presently in use is attributed to chance, divinities, foreign origin, and
change over time that produces deviations from an unrecoverable original.19 In
such cases, mere identifications of source—however vague—substitute for rational
accounts of procedures. Plato insists that the genuine expert (technikos, 42637)
must be able to provide the requisite logos, and he labels the latter three sources
noted above (divinities, foreign origin, and change over time) as mere excuses
(ekduseis) for an inability to do so (42631-4). The logon didonai requirement is
also at issue early on when Plato takes steps to ground such an account in his
remarks on Form-based construction (389—390); in what follows, however, he
indicates the marked disparity between such a description and any process of
naming yet undertaken. In the Gorgias, rhetoric's occupation with the produc-
tion of pleasure makes it impossible for its practitioners to account rationally for
their activity, and this shortcoming plays an important role in Plato's denial of
its techne status (46462-46537). Plato's emphasis in the Cratylus on the fact that

18. Kahn (1973, 157) identifies this point as "the decisive shift." Although this extension is
certainly important, in the dialogue's broader context it constitutes simply the second stage in a
three-stage progression. Gaiser (1974, 65) omits mention of stage i.
19. Regarding divine origin, see 39^-392^ /fi6c (the dianoia of gods or human beings, or
both), 4380. Concerning names' foreign source, see 40^-410^ 4163; worth noting, moreover, is
42ic-d, where mention is made of both foreign origin and deviation from an unrecoverable native
original. On chance (tuche) as source, see 39469, 39565.
88 THE ANCIENT QUARREL

practitioners of naming cannot exhibit any "understanding" they might have in

a logos reinforces his contention that they in fact lack knowledge altogether; the
result is a significant blow to the contention that naming is a tec/me.20
Regarding the subject-matter condition, Plato concludes that previous onoma
construction focused ultimately on the realm of appearance, with no appeal to
the Form of Name or to natures, properly construed.21 He makes clear that par-
ticular needs, beliefs, fears, and desires—whether individual or collective—pro-
vide motives or occasions for naming and that acts of production occur with the
empirical world in view. These factors, taken together, determine what results
from efforts at construction. Yet this is precisely the sort of activity to which Plato
denies techne status in his treatment of shuttle construction (38935^4), where
he stresses that construction undertaken with anything other than Forms in mind
categorically disqualifies any pretension of the activity in question to the rank of
techne narrowly construed, leaving it capable of techne status only in a generic
and undistinguished sense.
One may also assess the situation from the vantage point of products' em-
ployment. Indeed, since the constructor's supposed task is to produce onomata
with a view toward their use to reveal natures, one must evaluate them to see
how well they achieve this end. In the etymological section, it becomes evident
that, whether one treats proper names or general terms, each analysis is merely
aspectual and reflects the interpreter's own biases.22 Onomata themselves are
moreover, subject to change.23 The important topic of individual names' ulti-
mate lack of utility in the disclosure of natures will receive additional treatment
in the following sections of this chapter.
Despite such wide-ranging criticisms, for some time Plato persists in his simu-
lated adherence to incompatible claims. Most notably, he continues to insist that
the analysis of words' constitution is nature revealing (42861-3) long after the
ground of the naming process has been identified as mere belief (4iib-c). One
of these claims must be renounced: Plato's lengthy investigation of naming, which
illustrates its failure to meet those techne conditions presented in the Gorgias,

20. Interestingly, the alleged technai of rhetoric and naming are mentioned together (42534-
5) in a passage leading up to Plato's remarks on the logon didonai requirement (425d-426b).
21. While some have viewed Plato as positing multiple Forms of Shuttle, and hence of Name,
on my interpretation the text best supports the claim that there is a single Form of Shuttle that is
implemented in different ways in particular cases. For the Form of Bed (Rep. 596) described a
"an idealized functional property," or that which "provides the functional aim," see Moravcsi
1992, 67. On the present interpretation, the Form of Name plays an analogous role in the rele-
vant linguistic arena.
22. Examples are found at 395b2-c2, 39gd2-4oob7, 4oob8-cio (where explicit mention is made
of the fact that various interpretations are possible), 4oibi-ei, 40^11-40333, 40335-40^4, 4<34d8-
40633, 4o6bi-6, and 41231-4 conjoined with 43732-8 (where Plato offers contrasting analyses of
episteme).
23. See 4i4b-e (where the point of departure is an analysis of the term techne), 4183-4191:1,
4203—b.
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS INQUIRY 89

paves the way for his rejection of the former contention at the close of the dia-
logue (438-440). Having completed his account of this practice's fundamental
shortcomings, Plato cannot but reach a negative conclusion regarding its techne
status, hence also its allegedly close ties to dialectic.24

3. Nature and Convention:

The Cratylus Resolution
While the techne issue has been of relatively little concern to interpreters, much
attention has been devoted to Plato's stand on the nature-convention dispute that
shapes and permeates the Cratylus discussion. It is not possible, in my view, to
give a simple answer to the question of what his own position actually is. Rather,
as I will suggest, in order to do justice to the complexity of Plato's deliberations,
one must offer two distinct responses to this query.
Plato's treatment of the term sklerotes ("hardness") and the onomata of num-
bers (434c~435c) signals his embrace of the "conventional" pole of the Cratylus'
nature-convention dichotomy; that is, he maintains that one judges the correct-
ness of words' constitution by appeal to custom (ethos) and convention (suntheke).
At the outset of the dialogue, Plato mentions two alternatives to natural correct
ness, which he has Hermogenes conflate (3840-3856, especially 385d-e). On one
view, any onoma that anyone assigns to any entity is, simply by virtue of that fact,
an appropriate one for that entity. According to the other, correctness is deter-
mined by reference to the practices of particular linguistic communities, for
instance, poleis. The distinction is thus between a relativistic conception, which
does not assume agreement, and one based on convention. On the present inter-
pretation, Plato espouses the second of these alternatives.25
The example of sklerotes makes clear that the issue for Plato is not the revi-
sion of words' present constitution. The point of contention is the fact that lambda
"signifies the opposite of hardness" (TO evavtiov 5r|Xoi <TK?ir|p6'cr|TO<;, 434dy-8)
When Cratylus suggests that the lambda might be replaced by a rho (434di2),
this proposed change is rejected by Socrates as unnecessary since the presence
of letters representing opposites in no way inhibits communication. The possi-
bility of success in this endeavor does not depend on terms' descriptive content
revealing—however imperfectly—the natures of their referents. Rather, it is based

24. This outcome pertains exclusively to naming as here construed. Plato never involves any-
one besides the philosopher, in contrast, when making his own innovative proposals about orthotes
onomaton in the Phaedo and Republic 5. This exclusivity of function remains intact even in the
Sophist and Politicus, where, in connection with the Method of Division (diairesis), the construc-
tion and excision of terms assume a position of some importance.
25. For a different view, according to which the outcome of Socrates' early discussion with
Hermogenes is the rejection of conventionalism, see Barney 1997.
90 THE ANCIENT QUARREL

ultimately on the fact that certain combinations of letters are sanctioned by con-
vention as the onomata of particular entities; that is, signification (deloma, 435bz)
is accomplished by such combinations of like and unlike letters as happen to be
established. Plato focuses on the presence of letters representing opposites because
this indicates most clearly that likeness is inessential. Additional support for this
conclusion about convention's governing role lies in the fact that the word sklerotes
also contains letters that are irrelevant, and as such undesirable, from the perspec-
tive of likeness (consider the letters eta and omicron as analyzed at 427^.26
Plato turns next to the onomata of numbers: "Or, if you like," Socrates says to
Cratylus, "consider the case of number. How do you think that you could get
names bearing a likeness relation to each number unless you allowed agreement
and convention to have authority over the correctness of onomata? (rcoSev olei
E^EIV ovojiaia otioia evi eKaotcp TWV ctpi0|j,a>v ETtEveyKeiv, sav (if) eaq ti THV
crrjv 6(ioA,oyiav KCU cruv6r|Kr|v icupot; e%etv tcov ovoumoov 6p06Tnto<; Ttepi;)"
(435b6-C2). This second example reinforces Plato's claim about convention's
governing role. While one might wish to contend that the presence of ti ("some")
in b8 suggests a limited conclusion with respect to terms' constitution, namely,
that Plato is circumscribing the range of convention's application, this view is
called into question by the particular example chosen: the onomata of numbers
do not presently indicate by likeness, and as Schofield (1982) has observed, they
could be made to point toward essences only if complete authority were given
to convention.27 Plato's claim is thus not that convention makes a limited con-
tribution to fitness when the constitution of onomata is at issue but instead that
correctness is determined thereby. Moreover, since entities in the realm of
mathematics and of values constitute the core group of Platonic Forms, his choice
of example to reinforce the point about convention is surely not coincidental.28
Yet Socrates then seems to praise the desirability of a likeness-based approach
to fitness (435C2-dr). Anagnostopoulos (1972, 736) uses this passage to support
the view that "Plato preferfs] a language consisting of natural names" (cf. 1973,
332), and Baxter (1992, 186) cites 43JC2-3 in support of the claim that Plato
"believed in the desirability of a language that was as mimetic as possible." In
the view of Schofield (1982,80 [with n. 6]), in contrast, Socrates' remarks at435C2-
di are meant ironically. Such "praise" would be unsurprising, even expected, based

26. On the constitution of sklerotes, see Anagnostopoulos 1972, 732 (with n. 44).
27. "If the names of numbers are to disclose by resemblance their essences, then such disclo-
sure can only be effected by a positive use of convention. For example, we can disclose the differ-
ences between i, 2 and 3 through their names only by some purely conventional device such as
giving the name of i one syllable, that of 2 two syllables, that of 3 three; and it will have to be
agreed by convention that they are designed to signify numbers in the first place. This is the one
place in the dialogue where we glimpse the idea that representation is not a natural relationship,
but is itself subject to convention" (Schofield 1982, 79).
28. Notably, while mathematics is in the foreground here, values take center stage in Plato's
closing remarks about onomata and reality (438-440).
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS INQUIRY 9!

on Socrates' earlier "partisanship" of the natural-correctness theory. Even as he

offers this seeming encomium, in fact, Plato signals his own reservations. In this
regard, Schofield (80 n. 6) notes Plato's "emphatic reiteration of KOTO TO SWOTO
('so far as is possible . . . / 'z/it were possible . . .')" (435C3 and 7, respectively).
Another relevant indication is Plato's use of isos ("probably," "perhaps"; cy) in
the latter, conditional, remark about possibility. Further reflection, moreover,
reinforces the view that this "expression" of residual attraction is insincere.
Indisputably, having the right ontology is, according to Plato, a necessary condi-
tion for making proper determinations of terms' correctness. Would it be, however,
a sufficient foundation for a constitution-based approach to fitness? The analyses
of letters (426-427) privilege empirical considerations (e.g., what is involved in or
attends human sound production), which cannot be used to specify the nature of
abstract entities. Unless one could somehow produce another basis for the account
of letters, this consideration is decisive straightaway. Moreover, even if one did
succeed in offering an alternative foundation, that is, one not privileging empirical
phenomena, agreement would still undergird conclusions about which sounds
revealed which elements or aspects of reality (i.e., were "naturally" correct). This
need not be regarded as an anachronistic observation: if one accepts Schofield's
interpretation (1982,79) of the case of number, then Plato is quite aware that assign-
ments dubbed "natural" in this way would actually be governed by convention.
Etymology, in turn, provides a satisfactory framework for determinations of fitness
only if analyses of words' semantic constitution yield insight into their referents'
natures. While those who had the right ontology would not need to rely on or seek
"instruction" therein from words' descriptive content, individuals who, lacking this
knowledge, turned to words' constitution for guidance would simply find confirma-
tion there of that view of reality which they had brought to the enterprise. Either
way, individual words' semantic constitution does not itself reveal the natures of
their denotata. Plato's use of dual etymologies ofepisteme (4123 and 4373) to illus-
trate the fact that words are ambiguous (amphibolon, 43733) makes this point strik-
ingly clear: it cannot be a matter of coincidence that he selects this term to under-
score the inadequacy of etymology as a source of understanding!29 One might strive
to "eradicate" indeterminacy, once again through fiat, by declaring which semantic
breakdown of individual terms is correct—in the case of episteme, for instance,
that according to which the soul stops at things (histesin . . . epi tois pragmas!,
43734-5). This "solution" via pronouncement, however, is not available to a

29. For a different interpretation of the two analyses, see Sedley 1998, 151. In his view, the
objection that onomata can be interpreted so as to mean any number of things depending on
one's agenda would not likely "impress Plato much" (144). He cites as support for this claim 4i4cl-
e, which involves the addition and subtraction of words' components. While the passage indi-
cates that these changes may promote variation in analyses of an onoma over time, it does not
address the point that even if no shift occurs, interpreters can still decompose a word in multiple,
even contradictory, ways, especially if they are granted the sort of latitude that Socrates increas-
ingly allows himself as the etymological inquiry proceeds.
92 THE ANCIENT QUARREL

genuine naturalist. The point here is not that onomata specifically, and logoi
generally, are not useful to one who is, or would be, a philosopher. They are,
indeed, crucial avenues through which the exploration and articulation of real-
ity take place. As Plato stresses in the Theaetetus, even thought must be under-
stood on the model of dialogue, namely, as a kind of conversation that occurs
within the mind itself (1896-1903). As media of inquiry and of the presentation
of results, onomata and logoi obviously have descriptive capabilities. This is a
far cry, however, from a stance according to which the descriptive content of in-
dividual onomata itself grants insight.30
The distinction between the Cratylus and other dialogues as regards etymol-
ogy does not lie in its prominence in the former and total absence in the latter.
Plato offers etymologies elsewhere, though their presence serves a function quite
different from that associated with the natural-correctness thesis. Qua a literary
device present in philosophical discourse, such derivations belong to the same
genre as Platonic myths and allegories, but they differ sharply from them at the
species level. Most importantly, literary devices like those myths at the end of
the Gorgias, Phaedo, and Republic both reinforce or recapitulate points made
elsewhere in these dialogues and treat important ideas that are arguably not
amenable to discursive formulations. Thus, for example, the myth offered in the
Phaedo (1076-1153) both reprises observations previously made about reality and
appearance and makes points about the soul's judgment that would be conveyed
far less successfully by other means.31 Regarding etymologies, in contrast, one

30. According to Gonzalez (19983, chap. 3), although the Cratylus rejects the idea that ety-
mologies yield insight into phuseis, for Plato "a name is natural insofar as its use reveals the
true nature of the thing signified" (92; cf. 87, 89, 90). It is not evident precisely how, according
to Gonzalez, individual onomata could be nature revealing in a nonetymological sense. This
unclarity about the content of Gonzalez's view stems, at least in part, from the fact that he moves
without distinction between the level of names specifically and that of language more gener-
ally. Concerning the latter see, e.g., his comment (93) that the Cratylus "provides us with the
insight that language, as presupposing in its function the existence of stable natures, can be
used in such a way as to make these natures manifest." Since the Cratylus concentrates on the
plane of onomata, and since Plato is well aware of the distinction between the level of indi-
vidual names and that of language more generally, saying that language can be employed to
this end is not obviously the same as saying that onomata can accomplish it.
31. While I concur with McCabc (1992) that the myths are incomplete on their own, I can-
not agree that they function as challenges to central claims made in the arguments, Plato's goal
being to compel us to weigh the opposing views. According to Elias (1984, 36), "the myths are
Plato's poetry, designed not only to escape the criticisms levelled at the other poets but in the
manner of their telling to supply the defence he has called for" (cf. 238). On his interpretation,
myths assume the role of "axioms" in Plato's philosophical system because he recognizes that
dialectic "cannot attain its goals by demonstrating the truth of first premisses" (37). Poetry in the
form of myths will thus be required not only by people who are less than fully dominated by rea-
son but also by philosopher-rulers themselves (37-38). Elias does not, in my view, distinguish
adequately between poetry proper and poetic devices or argue convincingly that this much philo-
sophical weight must be given to the latter. For a critique of the view that Plato's defense of
poetry takes the form described by Elias, see Kraut 1987.
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS' INQUIRY 93

may agree with Meridier (1931, 18-19), wno' having noted their employment
outside the Cratylus, stresses that one has no reason to infer from their presence
in that dialogue or elsewhere that Plato was making serious philosophical use of
the procedure. At best, Plato's playful associations outside the Cratylus serve in
the first of those capacities noted above, namely, to reinforce points made by
him on independent grounds. When explaining the origin of the simplest form
of polls in the Republic, for example, Plato focuses on the issue of need. Human
beings have more needs than they can meet individually. From the fact that many
people (polloi) must therefore come together to ensure that all needs are met
arises the name polls as the designation of their community (369^. The etymol-
ogy, which is grasped properly only in its surrounding philosophical context,
privileges nothing of import that Plato's exposition does not otherwise bring to
light. The function of this analysis, like that of "Hades" in the Phaedo (see chap. 2,
n. 41), is to lend emphasis to a point that Plato wishes to make relying on other
considerations. In contrast to central myths and allegories, such etymologies,
while arguably of merit as literary flourishes, are treated as inessential to the
conduct of philosophy. Not surprisingly, then, as Plato's differing analyses of
"Hades" in the Cratylus and Phaedo make clear, far from trying to revamp the
practice of etymology so that analyses succeed in being naturally correct, he
simply allows contextual factors to guide his choice of source-onomata.32
Plato's discussion of nature and convention at Cratylus q^c-^^c is regu-
larly assumed to apply only at the level of terms' constitution. As we saw, on this
issue Plato comes down firmly on the side of convention. This does not, how-
ever, exhaust the scope of application of his remarks at that point in the dialogue.
In my view, Plato is operating on two distinct levels. This, I believe, explains a
curious fact about the passage, namely, that while the content of the examples
chosen illustrates the view that convention has the role specified above, Plato's
remarks include qualified formulations. Thus, Plato's discussion of the letters
comprising sklerotes leads him to conclude that convention determines a name's
correctness (43538) since like and unlike letters, combined in whatever way they
happen to be by custom and convention, do the job of signification (43538-10).
Distinguishing custom (ethos) from convention (suntheke) will do no good, since
one can simply rephrase the point to be that custom, not likeness, determines

32. Plato's handling of ems moves along similar lines: for relevant passages, see Cra. /poa-b;
Phdr. 2380. For differing analyses of himeros ("desire"), see Cra. 4196-4203; Phdr. 2510 (cf. 2550). By
giving his audience unmistakable interpretive cues, Plato avoids that scenario, deplored in the
Cratylus, in which people simply project their own assumptions onto their constructions of words'
descriptive content and hence find automatic confirmation there of whatever beliefs, however defec-
tive, they already happen to hold. As I have observed elsewhere (see Levin 2000, 283 n. 6), one in-
dication that Plato is not entirely serious in Socrates' second logos is that he treats etymology as
evidence in the "argument" about mania (Phdr. 244b-d; T.68e nfyv ci^iov £7UH.apTUpaa6ca, b6) although
he rejects the view that etymological analyses have independent philosophical value in the Cratylus
(434-440). For further discussion of ties between the Phaedrus and Cratylus, see Levin 2000, 283-84.
94 THE ANCIENT QUARREL

the correctness of terms' constitution (aio-bz), and that, Plato reiterates, "sig-
nifies by both like and unlike" (bz—3). Yet instead of closing the discussion of
sklerotes on this note, as one would expect, Socrates ends by saying that we
must therefore suppose convention and custom to contribute something (ti
sumballesthai) to the disclosure of meaning (435b5-6). Similarly, despite the
import of the number example, Plato interprets it to show that "some authority"
(ti kuros) must be granted to convention (435b8-ci). Rather than signifying a
partial allegiance to naturalism with respect to the constitution ofonomata, such
qualifications suggest that, even as Plato rejects the pertinence of "nature" when
one judges the fitness of terms' constitution, he leaves room for the possibility
that it could operate on some other level.33 The Cratylus' scope is thus limited,
insofar as it concentrates on just one dimension, albeit a crucial one, of what
underlies or is involved in determinations of fitness. Yet Plato's hint at 435b-c
that what he has said thus far does not constitute the whole story of how nature
and convention stack up against one another, even interrelate, is picked up, as
we will see, in the dialogue's closing remarks (438-440). These considerations,
taken together, indicate that Plato has not yet exhausted the topics of nature and
convention insofar as the two concepts pertain to judgments of terms' fitness.

4. The Dialogue's Conclusion

In the Cratylus, terms denoting key elements of Plato's philosophical system are
subject to the same type of analysis as other onomata: central among them are
psuche (399d-4oob), phronesis (4iid),34 noesis (4iid-e), sophrosune (4116-4123),
episteme (4123,4373),35 sophia (412!}), agathon (4120), dikaiosune (4120), dikaion
(412c-413d), andreia (4136-4143),36 techne (4i4b-c), arete (4i5c-e),37 kalon

33. Cf. Silverman 1992, 64-65: "Convention contributes something to the correctness of
names. This implies that nature also contributes. However, the force of this argument . . . sug-
gests that nature is not to be parsed as Cratylean likeness."
34. With regard to phronesis outside the Cratylus, see, e.g., Rep. 433b8, where the term ap-
pears in place of sophia in Plato's list of the four excellences (aretai) (b8—ci). For phronesis in the
sense of theoretical understanding, see Phd. 6663, 6832, 7, 69310, bj—4, 76012, ygdy, 11408 (cf.
Gallop [1975] 1988, 102). For phronesis, episteme, and sophia as synonymous, see Symp. 20232-9.
35. Episteme is precisely what turns out to be lacking in the practitioners of what is alleged 3t
the outset to be a techne. Interestingly, the noun itself is not used in the dialogue with reference
to these individuals; Plato reserves for it the role of illustrating words' deep-seated ambiguity.
36. Plato's treatment here of andreia completes the core list of aretai of interest to him (the
others being sophrosune, sophia, and dikaiosune); these are the four he handles in the Republic.
Agathon mentions the same four in his encomium of Eros at Symp. ig6b-e though they are not
discussed there based on Plato's own views; for observations on Agathon's misinterpretation of
each arete, see Nehamas and Woodruff 1989, 34-35.
37. According to Benardete (1981, 138), "the thirteen names from 'courage' to Virtue' form a
group," in which "'art' [techne} and 'device' [mechane] are intrusive elements." In my view, techne
in particular is not "intrusive" insofar as there is for Plato a necessary correlation between one's
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS INQUIRY 95

(/fi6b-d), eros (4203—b), aletheia (42ib), on and ousia (42ib-c).38 Moreover,

Plato treats the term hedone (4:^7^1), which cannot receive here the sustained
consideration provided in the Gorgias. Last but not least, he subjects the term
onoma itself to analysis (42137-^1), which is noteworthy given the fact that it is
precisely onomata whose distinction as "naturally correct" is at issue in the
Cratylus. From within this framework, Plato has no dependable means of privi-
leging the aforementioned terms and their referents in the way he believes they
deserve.39
This structure also provides no reliable way for Plato to draw the proper dis-
tinctions between certain onomata analyzed in the Cratylus, specifically of course
their referents, where the referents in question form contrasts playing important
roles in his metaphysics, epistemology, ethics, and moral psychology. I have in
mind, for instance, the dichotomy between soma and psuche (jggdi-^oocio),
of which he makes so much elsewhere —it underlies the classificatory scheme
of the Gorgias and has a central role in the Phaedo—as well as that between
arete and kakia (41539-62). Also relevant here is the contrast between kalon and
aischron (4i6aro-dn), which is central to the Symposium, and that involving
aletheia and pseudos (42^1-7).40 In addition, Plato's etymological discussion
includes the terms epithumia and thumos (41x^8-62), which figure importantly,
and are distinguished, in Book 4 of the Republic (4396-4403).41 Even more notable
is the fact that Plato provides the terms of the contrast between episteme, noesis,
phronesis, and sophia, on the one hand, and doxa, on the other—a dichotomy
essential to his epistemological reflections in the Republic and several other
dialogues.42 Central here are terms denoting concepts and distinctions that play
crucial roles in the Phaedo, Republic, and Symposium and that, in many cases,

conception of the best life, as reflected in the caliber of those activities (praxeis) in which one
engages, and how one is situated with respect to virtue and vice (for the analysis of kakia, see
4155). While the precise nature of that relation does not emerge here in the context of etymologi-
cal inquiry, Plato's combination of analyses reflects his conviction that the issues are related.
38. One key measure of the futility, even danger, of this approach is found in the etymology
conflating einai and ienai, according to which "on and ousia are ion with the iota removed. This
accords with the truth, for being signifies going, and not-being (ouk on) — as some even name
it—not-going (ouk ion)" (42^7-02).
39. The fact that the group of onomata containing virtually all of these terms is situated last
in Plato's sequence of analyses is indicative of their referents' importance from his perspective.
The formation of this group, however, is merely a preliminary stage in a process that aims to pro-
vide an account of those entities' natures and relations, and that account cannot be provided by
analyses of terms' descriptive content.
40. See also Rep. 47569, where Plato states explicitly that kalon and aischron are opposites.
41. There Plato labels the lowest element of the soul to epithumetikon ("the appetitive," 43965;
cf. d6-8) and uses the noun epithumia to represent appetite in a general contrast with the soul's
spirited aspect (44035-6). He also maintains that each of the three elements has its own desires
(epithumiai) (58od8).
42. Plato treats the term doxa at 4201)7-9.
96 THE ANCIENT QUARREL

are subject there to sustained philosophical analysis. As noted above, it is surely

not coincidental that episteme is the term Plato employs to deliver the coup de
grace to the etymological approach to fitness. Since the claim under scrutiny is
that etymological analysis yields understanding, it is quite significant that utiliz-
ing the procedure provides no dependable guidance as to what the nature of
understanding itself actually is.
Although it is ultimately convention that explains the correctness of terms'
constitution, one's application of terms to entities will be fundamentally misguided
if one is not aware of what does have a natural foundation. In Plato's view, before
addressing the topic oforthotes one must consider which entities are fundamental
and which derivative. In the closing pages of the Cratylus (438-440), he addresses
most directly the question of what having a "nature" involves. Plato's ousiai com-
prise neither mortal or divine individuals nor particulars generally. From his per-
spective, nothing that is not unchanging and unchangeable —and therefore noth-
ing that is native to the sensible realm or to the world of capricious, changing
gods—can properly be said to have a nature and hence to be worthy of investiga-
tion in its own right (439C7-44od4).43 Plato contrasts here "the beautiful itself and
good itself and each existent with this status" (amo KaA-6v Kcd aya&ov Kai ev
EKCXOTOV TCOV ovxcflv oikw)—about which Socrates is said to have frequent dreams—
with "some beautiful face or another of such things" (Ttpooomov n KaA,6v r\ ti trav
TOiomwv) (439c8-d4). This mention of the reality-appearance dichotomy calls to
mind those numerous passages treating the distinction in the Phaedo, Republic,
and Symposium, where language of "nature" (phusis) is used explicitly of Forms.44
Plato notes, in addition, reality's function as explanans (Cra. 439a-44od).

43. This is not to say, of course, that Forms as Plato depicts them share no attributes with
divine individuals as treated by the literary tradition; for example, neither type of being is subject
to perishing. In addition, both taking as a backdrop and departing from this familiar conception
of divinity, he applies the term "divine" (theion) to the realm of Forms (Phd. 8obi). Apropos of
the claim about particular entities' lacking natures of their own, a few words are in order about
the Phaedo's handling of fire and snow (103-106); for illustrative purposes, I concentrate on the
former. While fire is expressly identified as a Form in the Timaeus (jibS), it is not clear from
what Plato says in the Phaedo whether he yet considers it to be a Form rather than some other
type of natural unit. Even if fire turns out not to be a Form, however, it is still a property of some
sort. Moreover, the properties of fire and heat are necessarily related. In a given case, the connec-
tion is as follows: Of any K such that x is fire, x is hot. One has here an instance of class subsumption
whereby the class of fiery things is a subset of the class of hot things. On neither interpretation,
then, can any particular fire be said to have a nature. For an argument that the fire and snow of
the Phaedo are the ordinary physical entities, see Nehamas 1973, 482-90.
44. For a different interpretation of Cra. 4390—44od, see Irwin 19773, 2; cf. Dalimier 1998,
56-60. I cannot agree with Luce's use of Plato's mention of dreaming (oneirottein, 43907) to sup-
port the claim that the Cratylus presents the theory of Forms at a distinctly earlier stage than do
the Phaedo and Republic (1965, 25-27). Gaiser's alternative interpretation (1974, 53) seems to me
to be the correct one: "Thereby is the weight of this thesis not diminished; the idea is rather that
its foundation is not articulated here." On the reference to dreaming, cf. Meridier 1931, 136-37;
Baxter 1992, 177.
 THE STRUCTURE AND OUTCOME OF THE CRATYLUS INQUIRY 97

In addition to sketching his conception of reality, Plato hints at his own

approach to the issue of correctness. First, he issues an exhortation:

Let us seek that beauty itself, not whether some face is beautiful, or any other of
such things, since all these things seem to be in flux. But don't we assert that the
Form of Beauty always remains just as it is? (43gd3-6)45

Then, immediately following this remark on the proper orientation, he issues a

contrasting statement about individual sensibles:
Can we appropriately call something "beauty" if that thing is always withdraw-
ing, . . . or is it not in fact necessary that at the very same moment we speak of it,
it straightaway changes into something else and withdraws, and in no way remains
in the same state? (439d8-u)

Plato suggests that individual sensibles' ontological deficiency precludes their

being properly (i.e., unqualifiedly) spoken of as "beautiful." His statement at
439d8-n, interpreted against the backdrop of d^-6, implies that one cannot make
judgments of fitness of the same type about Forms and particulars. These final
comments move in the direction of an alternative theory of appropriateness in-
sofar as they point both to the existence of a hierarchy in the application of cer-
tain onomata and to the important role of ontological considerations in any given
judgment of correctness.46 Although what Plato says here could serve as the germ
of, or point of departure for, a positive account of fitness, it is itself a far cry from
the presentation of such a theory.
Having merely broached the subject of appropriateness from his own philo-
sophical perspective, he turns in the dialogue's final lines to the matter of knowl-
edge and requirements on its proper objects. Plato merely sketches his meta-
physical theory in the Cratylus: he invokes the reality-appearance dichotomy
without treating it at length and does not introduce the notion of participation,
which is central to his treatment of appropriateness. While I concur with Silverman
(1992) that Plato aims to show how a given onoma can refer primarily to a Form
and derivatively to its participants, I therefore cannot agree with his claim that "the
key to the [Cratylus] is to see that it is directed at answering how a given name can

45. As Kahn (1996, 365) points out, this contrast between the Form of Beauty and a beautiful
face calls to mind Symp. 2113.
46. While I do not concur with those who find a full-fledged positive stance in the Cratylus,
I thus also cannot endorse the contention of Steinthal ([1890] ig6r, 112) and Schaarschmidt (1865
344) that the dialogue's outcome is wholly negative.
98 THE ANCIENT QUARREL

be both an 6vo(j,a and an eponym" (27). In my view, with the exception of Plato's
remarks at 43901, a concern with this twofold application of terms emerges, not in
the Cratylus, but rather in the Phaedo, to which I turn in chapter /)..47
Although the Cratylus' own emphasis is largely critical —insofar as it contests
inadequate conceptions of naturalness and appropriateness—it thus serves as a
crucial locus of Plato's theorizing; as such, it is far from having an isolated position
in his corpus, as some commentators have maintained.48 Most important for present
purposes, it is one of three middle dialogues that treat issues in the domain of
orthotes onomaton against a literary backdrop. As I will argue in the following
chapter, while the Cratylus denies the value to philosophical semantics of its liter-
ary antecedents, in the Phaedo and Republic 5 Plato develops his own conceptions
of eponymy and the use of functional terms, respectively, with their literary back-
ground in view. Early in the Cratylus, Plato alludes to eponymy (in his repeated
mention of "Scamander" near his discussion of "Scamandrius") and treats briefly
a key functional term, namely, hasileus ("king," 393-394). Here, however, this is
as far as his consideration of these notions extends. When, in contrast, Plato explores
what he takes to be more fruitful approaches to the handling of correctness, these
other frameworks take center stage. Chapters 2—4, taken together, thus offer a fuller
picture than before of the scope and depth of Plato's engagement with literature
in the development of his views about orthotes onomaton—the central arena in
which his concern with the philosophy of language manifests itself in those middle
dialogues that articulate the theory of Forms. Moreover, as we will see, there is no
indication in the Phaedo or Republic that by turning away from etymology, and
toward eponymy and the application of functional terms, Plato is settling for sec-
ond-best. On the contrary, these frameworks, suitably revamped, are presented as
optimal contexts in which one may treat questions of appropriateness given the
metaphysics sketched in the Cratylus and elaborated in these other dialogues.49

47. Cf. Kahn's remark (1973, 175-76) that "the Cratylus never mentions participation, and
the fuller ontological framework of Plato's theory of names is merely hinted at," though, in my
view, this hint does not consist simply in Plato's "mention of the dialectician as the user and judge
of names" (176 n. 29). As I argue in chapter 4, far from embodying opposed stances (Schaarschmidt
1865, 346-47) or being unconnected in their philosophical content (Bostock 1986, 160), the two
dialogues' relation is strongly complementary. In stressing the Cratylus' link to the Phaedo with
respect to orthotes onomaton, I differ from Kahn (1996), who groups the Cratylus with the Sym-
posium and Phaedo in terms of its date of composition, yet tics it most directly to the Theaetetus
and Sophist with respect to its treatment of naming and other themes (364; cf. 366).
48. For the Cratylus as occupying such a position, see Fowler 1926, 4; fowett 1892, vol. r, 254;
Wilamowitz 1959, 224.
49. For yet another manifestation of Plato's concern with orthotes, one may consult the Sophist
and Politicus. Notably, even here, where onoma construction is described as one of the dialectician's
responsibilities, issues of constitution never govern the selection of onomata in the way required
by that "model" alleged by some to be developed in the Cratylus. Exploration of Plato's view in
the Sophist and Politicus would take one well beyond the scope of the current project. For brief
remarks on his handling of appropriateness therein, see Levin 1995, m-r3.
 4
CONCEPTIONS OF
APPROPRIATENESS
Plato's Revision of Literary Usage in
the Phaedo and Republic 5

i. Pouring New Wine into Old Skins: Reflections

on Literary and Platonic Eponymy

1.1. Introduction
While the Cratylus targets what Plato views as a faulty approach to orthotes, the
Phaedo presents his own theory of the basis on which onomata are correctly
assigned to their referents. As has long been recognized, this account centers on
the eponymy, or "named-after," relation. Thus, at a crucial point in the dialogue,
Plato notes that Forms exist and that "the other things, participating in them,
are named after the Forms themselves" (xomcov TaXA.a (j.etatax|j,pdvovTci omxcov
totraov Tf|v ejccovuniav ia%evv, 10202-3). Subsequently, he articulates a general
and fundamental contrast between things that have opposites, which are named
after those opposites (eicovoiid^ovTec; amd TTJ eKeivcov ejicovunia), and "those
opposites themselves from whose inherence in them the things named receive
their appellations (e%ev Tr|v e7io>vu|ii.av)" (ic>3b6-ci).
The framework of eponymy plays a central role in Plato's reflections at this
juncture because it provides the semantic correlate to the notion of participa-
tion, which is itself crucial to his metaphysical theory. Platonic eponymy under-
scores dependencies. In so doing, as we will see, it does far more than bring out
mere historical linguistic ties between onomata. Rather, it both reflects and draws
attention to ontological and valuational dependencies involving Forms and
members of the pertinent classes of entities in the spatiotemporal realm. As in
the case of etymology, so too with eponymy, Plato is quite concerned with the
issue of whether, and under what conditions, onomata are properly assigned to
their referents. While, as I have argued, Plato rejects etymology as a fruitful way
of approaching questions of appropriateness in a philosophical context, his atti-
tude toward eponymy is quite different. In the latter, Plato finds a semantic notion

99
1OO THE ANCIENT QUARREL

that, if properly interpreted, is well suited to mirror in linguistic terms the depen-
dent relationship that he believes to obtain between entities in the familiar world
and Forms—the exploration of which relationship is of pivotal concern to him
in the middle dialogues.
Although commentators have recognized the important role played by
eponymy in the context of Plato's thought, they have not analyzed its historical
sources in order to clarify what is distinctive about his approach.1 As previously
noted, Socrates, Parmenides, and Heraclitus are regularly identified as the key
philosophical sources of Plato's middle-period theory of Forms. While all three
help to shape the development of Platonic metaphysics and evince some inter-
est in issues of naming, none invokes eponymy as a solution to those linguistic
problems he does identify. 2 This semantic relation is important, not for Plato's
philosophical predecessors, but instead for the literary tradition. 3 Literary au-
thors' handling of the eponymy relation offers a heretofore unexplored prece-
dent for Plato's own systematic treatment of eponymy and his use of it to ground
judgments of appropriateness.4 An examination of literary eponymy will make
it possible to separate inherited presuppositions from those innovations that char-
acterize the Phaedo s new, technical use of the device.5 As the quotations from

1. For a recent emphasis on its importance, see Kahn 1996, 353.

2. Concerning Parmenides and the topic of naming, see DK 8.17, 36-39, 53-54; 9; 19. The
Diels-Kranz edition does not include in the fragments of Parmenides that quotation from Plato
(Tht. i8oe) for whose authenticity Cornford (1935) has argued. On this issue see McKirahan (1994,
157 [with n. 15]), who places the remark among the fragments, interpreting it to mean that "the
correct name for the one existing thing is 'to be."' For discussion of Heraclitus, see chapter 2, sec-
tion 3. Socrates' interest with regard to naming lies in determining whether or not what others
dub V (e.g., courageous or pious) actually is such (concerning this interest, see La. 1976; Chrm. ij$b;
Euphr. 6d-e). I endorse the view, which is widely shared, that early Plato is our most reliable
source regarding Socrates' philosophical views. For emphasis on the Apology's value as a check
on assessments of how far Plato may be trusted, see Vlastos 1971, 3-4. Since I take the aforemen-
tioned concern to be evident in the Apology, I consider myself to have adequate justification for
attributing it to the historical individual.
3. The present survey of literary eponymy, like that of etymology, encompasses Homer's Iliad
and Odyssey; Hesiod's Theogony and Works and Days; Pindar's epinician odes; the extant plays
of Aeschylus, Sophocles, and Euripides; and Herodotus.
4. For interpretations of the Phaedo that neither take account of the literary tradition's im-
portance nor explore, more generally, the historical sources of Platonic eponymy, sec Gallop [1975]
1988; Bostock 1986; Bluck 1955; Hackforth 1955; Robin 1926; Burnet 1911; Guthrie 1975, 324-65;
Friedlander 1969, 35-62, 471-77; Taylor 1960, 174-208; Jowett 1892, vol. 2, 157-266; Grote 1867,
152-205. For brief, general remarks about historical background, which do not mention the liter-
ary tradition, see Bestor 1978, 196-97.
5. I concentrate on the Phaedo because that is where Plato presents his version of eponymy
and indicates most clearly its importance. Whereas the literary tradition makes considerable use
of etymology and eponymy without distinguishing between the two in a direct or clear-cut way,
Plato distinguishes sharply between them and offers different assessments of their philosophical
value: in the Cratylus, as we have seen, he criticizes the literary tradition's use of etymology to
treat questions of appropriateness and denies its philosophical worth; in the Phaedo, in contrast,
he endorses for philosophical use a revised version of eponymy. Thus, the two dialogues' thematic
 CONCEPTIONS OF APPROPRIATENESS 1O1

literary texts provided below make clear, this study concentrates on the phenom-
enon of literary eponymy, by whatever terms it is introduced, rather than simply
on writers' use otepdnumos, eponumia, and eponomazein. In particular, I con-
sider how one might organize and describe literary practice, with special atten-
tion to those aspects of it on which Plato builds and with which he takes issue.
In light of this exploration of literary eponymy, I then turn to Plato's handling of
this semantic notion in the Phaedo, commenting also as necessary on the meta-
physical foundation of his approach.

1.2. Eponymy in the Literary Tradition

Before embarking on a discussion of literary eponymy, it will be useful to con-
sider briefly the relationship between it and literary etymology. Both involve trac-
ing one onoma back to another which is taken to be its source. In the case of
etymology, names often provide standards against which their bearers are judged.
Included here are assignments tied to features or aspects of individuals, notably
their attitudes and character traits or skills and capacities. Relevant in this con-
nection are, for example, the analyses of "Ctesippus" (Horn. Od. 20.288-90),
"Philoctetes" (Soph. Phil. 1344-47;c^-1413~33)> "Aphrodite" (Eur. Tro. 989-90),
"Dolon" (Eur. Rhes. 215; cf. 216-18), "Theonoe" (Eur. Hel. 8-14), "Demodocus
(Od. 8.43-45), "Thoas" (Eur. IT 32-33), and "Prometheus" (Hes. Th. 510-11 [with
the illustration in 535-60], Op. 54; Aesch. PV 85-87). As we saw in earlier chap-
ters, onomata are properly assigned if they disclose or describe bearers' natures,
and determinations of names' appropriateness are based on analyses of their
descriptive content. In contrast, in cases of eponymy, which are often treated as
commemorative, names do not function as standards to the same extent or in
the same way. While in certain instances one may imagine that questions of
"nature transfer" are relevant, as when leaders give their names to peoples, this
is typically not something that comes to the fore. Moreover, where one entity is
said to be appropriately "named after" another, this occurs without the intro-
duction of the issue of semantic constitution. A final point is that in the case of
etymology the primary terms themselves are not subject in turn to analysis. Simi-
larly, with regard to eponymy, where the source terms are usually proper names,
the derivation of those onomata is generally not explored. Interestingly, where
the primary terms are analyzed, this transpires via etymology, as when the name
"Ion," whose bearer is supposed to have given his name to the lonians (Hdt. 7.94,
8.44.2; Eur. Ion 1581-88), is itself subjected to etymological analysis (Ion 661-

relation is strongly complementary: though Plato offers limited clues there to his own stance,
mainly the Cratylus clears the ground for those metaphysical and semantic theories developed
in the Phaedo. The Phaedo's relation to the Cratylus, where the literary tradition's pertinence is
more readily apparent, is strong* evidence in favor of its pertinence with regard to eponymy.
102 THE ANCIENT QUARREL

63, 802, 830-31). Similarly, the name "Athena," whose referent is treated as the
eponymous deity of Athens (Soph. OC 107-8; Eur. Ion 8-9, 29-30,1555-56), is
also etymologized (Horn. Od. 13.330-32).
Literary authors' use of eponymy breaks down into numerous categories and
subcategories.6 The first and most popular category comprises instances in which
individuals give their names to a wide range of entities. These might be other
individuals, for instance, a male relative, as when Pindar claims that Strepsiades
was named for his maternal uncle
Isthm. 7.24) and that Opous was named for his mother's father (Ol. 9.63-64;
iorovuiioq is used). Underscoring the same type of relationship, Eteocles tells
Creon that he will send the latter's son Menoeceus, "the boy who has your father's
name" (coi) mrcp6<; ameji(»vu(a,ov), to get Tiresias (Eur. Phoen. 769). Accord-
ing to Herodotus, in turn, Cleisthenes got his name from his grandfather,
Cleisthenes of Sicyon (e%(ov TO owo|j.a GOTO TO\> |a,r|TpO7iaTOpo<; TOU SiK-ucovioa)
6.131.1; see also 5.69.1), and Bubares and Gygaea had a son who was "named
Amyntas after his maternal grandfather"
8.136.1). Elsewhere, Aeschylus reports that Phoebus was named after the Titan
Phoebe (Eum. 8), while Herodotus remarks that Hippocrates named his son
"Pisistratus" after Nestor's son (5.65.4) and that Cimon's son Miltiades got his
name "from Miltiades the founder of the settlement in the Chersonese" (onto
TO\) OIKIOTEM Tt\5 XepoovT|(K>'U MilTia8ecfl, 6.103.4). Still other examples are
found at Homer, Odyssey 18.5-7; Euripides, Heracles 31; Herodotus 1.188.1,3.55.2
and 6.131.2.7
Often groups of individuals are in the recipient position, as when Homer says
that Erichthonius, son of Dardanus, begat "Tros, who was lord of the Trojans"
(Tproa . . . Tpcoeooiv civaKTa, IX 20.230). Elsewhere, Aeschylus claims that the
Pelasgians were named after their king Pelasgus (Supp. 250-53), and Herodotus
reports that the descendants of Targitaus' three sons, taken as a group, were named
"Scoloti" after one of their kings (TO\) paoiAeog ejioovunvrrv) (Hdt. 4.6.2). Dis-
senting opinions about the origin of the name Oepoai are expressed at Aeschylus,
Persians 8o8 and Herodotus 7.61.3 conjoined with 7.150.2. Moreover, both
Herodotus and Euripides identify Ion as the source of the name "lonians."
According to the former, when the Pelasgians occupied what is now called
Greece, the Athenians —a Pelasgian people—were called "Cranai." In Cecrops'
reign, they came to be called "Cecropidae." When Erechtheus came to power,
they changed their name to "Athenians," and upon Ion's becoming general, they

6. These categories do not do justice to the rich and meaningful use of eponymy in litera-
ture. They are intended simply to facilitate the organization of this substantial body of material
with a view toward the themes under discussion in the present study.
7. With respect to Hdt. 3.55.2, I have in mind at this juncture Archias' being named for his
grandfather. I consider the name "Samius," also treated at 3.55.2, below.
8. If line 146, whose authenticity has been challenged, e.g., by Robertson (1924, no) and
Broadhead (1960, 67), were in fact genuine, it would be pertinent here.
 CONCEPTIONS OF APPROPRIATENESS K>3

assumed the title of "lonians" (8.44.2; see also 7.94). Euripides says that Ion's
grandchildren will be called "lonians" based on the name of their grandfather
(Ion 1581-88). For other relevant cases, one may consult Sophocles, Oedipus at
Co/onus 58-65; Euripides, Ion 1575-78, 1590-94^ Herodotus 2.42.5, 4.149.1,
5.66.2, 5.68.2, 7.11.4 (a people and their land named after the conqueror Pelops),
and 8.44.2.10
In addition, individuals give their onomata to places or parcels of land. Thus,
Castor instructs Orestes to found a city that will take its name from him (eTtcovDuoi;
5e (J0"u 7r6A,i<; KeKW|O£Tcu, Eur. El. 1275), and a parcel of land is called after
Apis, who had previously rid it of certain plagues (Aesch. Supp. 260-61). Else-
where, Bacchus is said to have given his name to a parcel of land (iao8' e7t(6vt>]j,ov
yaq) (Soph. 0X209-11), and Herodotus identifies Archander, son of Phthius, as
the source of "Archandropolis" (2.98.2). Regarding the connection between
Athena and Athens, Burkert (1985,139) notes that "whether the goddess is named
after the city or the city after the goddess is an ancient dispute. Since -ene is a
typical place-name suffix . . . the goddess most probably takes her name from
the city." Such linguistic arguments notwithstanding, on several occasions Athena
is associated with Athens as primary name-bearer to nominatum (Soph. OC 107-
8; Eur. Ion 8-9, 29-30,1555-56). Pindar, in turn, treats several parcels of land as
having eponymous nymphs: Thebes (Isthm. 3, 7-8), Aegina (Pyth. 8; Nem. 7-8;
Isthm. 8), Libya (Pyth. 4,9), and Rhodes (Ol. 7). With regard to this subcategory,
one may also consult Homer, Iliad 20.215-17, 231-32; Pindar, Olympian 7.73-
76; Euripides, Ion 74-75, Orestes 1646-47, Andromache 17-20; and Herodotus
4.45.2-4 (vis-a-vis this passage, which pertains to the threefold division of the earth
into Europe, Asia, and Libya, cf. Find. Pyth. 9.5-8, 55-58, 68-69)."
Furthermore, sometimes natural inanimate entities (e.g., bodies of water) are
in the recipient position. For instance, Aeschylus foretells that the inlet of the
sea to which lo is driven "shall be called Ionian, a memorial to all men of your
journeying"
PV 840-41), and Pindar remarks that Heracles "called [the hill at Olympia]
Cronus' Hill, for in former times . . . it had lain beneath deep drifts of snow,
without a name (vcovuuvoq)" (OL 10.49-51; cf. Ol. 5.19). Also relevant here are
Aeschylus, Eumenides 689-90 and Prometheus Bound 299-300. Finally, the
recipient entities may be human constructions or practices, as when Herodotus
comments that "the Delphians call this gold and silver that Gyges sent the Gygean
Treasure, after the treasure's donor (em mi) dvaSevTOc; e7uovu|iir|v)" (i.i4.3),12

9. On balance, it seems preferable to place these two passages from the Ion here rather than
in that subcategory in which parcels of land are the recipient entities.
ro. While the preceding citation of 8.44.2 involved Ion as eponym, this reference pertains to
Cecrops as primary onoma-bearer. For additional passages from Herodotus, see 1.7.3 ( c f- 7-74 J )>
1
-94-5-7> i-iyi-5-6, 1-173-3 ( cf - 7-9 2 ). 7-62-1, 7-9°-91-
ir. For additional relevant cases, see Hdt. 4.148.2-4, 6.47.1, 7.178.2.
12. My translation.
1O4 THE ANCIENT QUARREL

and Electra refers to a feast named for Agamemnon (Soph. El. 282-85). Other
examples of this type are found at Pindar, Pythian 11.5-6; Euripides, Hippolytus
31-33 and Orestes 1008.
A second category consists of instances in which natural inanimate entities
(e.g., rivers) give their onomata to various types of entities. These might be indi-
viduals, as when Homer notes that Telamonian Aias killed "Simoeisius. . . whom
once his mother descending from Ida bore beside the banks of Simoeis when
she had followed her father and mother to tend the sheepflocks. Therefore they
called him Simoeisius"

II. 4.474-77). Groups may be the recipients of onomata,

as in Herodotus' report that the people living around the mountain called
"Atlas" were named "Atlantes" after it (em. TOUTOU TO-U opeoq oi avOpcorccn OUTOV
eTtcflvufioi eyevovTO, 4.184.3-4); with regard to this subcategory, one may also
consult Herodotus 7-75.2. Natural inanimate entities themselves may also play
this role. To cite just one example, Herodotus mentions the city Aegae on the
Crathis, which is never dry, "from which the Italian river Crathis has its name"
(COT' OTETJ 6 ev 'iTaXirj nom\ioq TO owc^a eo^e, 1.145).n 1° addition, places or
parcels of land are names' recipients, as when Pindar mentions a city called
after neighboring Mount Aetna (toij ejicowjuiav noKiv, Pyth. i.30-32).H
A third category consists of instances in which places or parcels of land are in
the primary role. Here the nominata may themselves be places. In this regard,
one may consult Euripides' Helen, where Teucer informs Helen of Apollo's
prophecy that he will found "and name New Salamis from my island home"
(ovo^ia vricacmKov ZaXajjlva GEUBVOV Tfjc; EKE! %dpvv naipaq, 149-50). In one
notable instance, a kind of horse is said to be called "Nisaean" because it comes
"from the great Nisaean plain in Media, where horses of unusual size are pro-
duced"
, Hdt. 7.40.3). The recipient enti
ties might be, in addition, natural inanimate entities; in this connection, the town
Therma on the gulf is said to have given that gulf its name (Hdt. 7.121.1). More-
over, individuals receive their appellations from places associated with noteworthy
events, as when Archias reports to Herodotus that "his father had been called
Samius" (TCO jimpi e$r) Zriiuov Towoiia Te6f|vca) to commemorate the heroic
death of Archias' grandfather in Samos (3.55.2); for another pertinent example,
see Hesiod, Theogony 198. Human constructions too could get their onomata on
this basis. Along these lines, Athena tells Orestes to name the temple he will build
at Halae after Tauris (ETICOVDIIOV yfj<; TompiKfjg, Eur. IT 1454) to immortalize
his sufferings and eventual deliverance. One may round out this discussion of
categories by mentioning that, fourthly, a parcel of land might be named for a

13. My translation. For other relevant cases, see Hdt. 4.52.1, 7.58.3, 9.51.1-2.
14. With respect to this subcategory, see also Hdt. 4.198.1.
 CONCEPTIONS OF APPROPRIATENESS 1O5

people. Relevant here is Herodotus' report that Darius gave those Barcaeans taken
as slaves a village in Bactria to inhabit, to which they gave the name "Barca"
4.204).
In addition to their use of eponymy per se, and as in the case of etymology,
writers offer eponymy-based assessments of names' correctness. Thus, Pelasgus,
for example, comments on the appropriateness of his people's name: "I am
Pelasgus, founder of this land, and son of Palaechthon earth-born. The Pelasgians
are fittingly named after me their leader (euoij dvaKTOi; e-UAoyox; EJICOVUIJOV),
and reap the fruits of this earth" (Aesch. Supp. 250-53). In another case, Herodotus
mentions a river called "Hypanis" that has its source in Scythia in a great lake
that "is called—very properly (6p0dk;) — the Mother of Hypanis" (4.52.1). Here
the primary entity, a body of water, gets its own onoma from another such body
that issues from it. Elsewhere, Herodotus wonders "why three distinct women's
names [Libya, Asia, and Europa] should have been given to what is in reality a
single landmass (|J.ifi eowrj yrj)" (4-45-2). The clear implication is that, in his
view, there is only one genuine entity available for naming. Given that fact, the
allocation of three names—based on an artificial division of this landmass —is
simply inappropriate. In an instance involving the acquisition of a nickname,
Homer tells of a certain beggar who arrived at Odysseus' home: "Arnaeus was
his name, for this name his honored mother had given him at his birth (EK
yeveifjc;); but Irus all the young men called him, because (OWEK') he used to
run on errands when anyone bade him" (Ocf. 18.5-7).15 In cases of name
changes and the acquisition of nicknames, one is justified in raising the matter
of fitness since it is precisely because the new or additional appellation is
appropriate that it is introduced at all. This individual is named "Irus" after
the female deity Iris, the divine messenger of Olympus. He would have been
incapable of serving as messenger, and hence of meriting this onoma, at birth
or during his early life; rather, the nickname is applied to him based on his
current performance of a type of action with marked parallels to that of the
primary name-bearer.
Judgments of desert may also be strongly implied, as when Oedipus invokes
Athens in prayer: "Hear me, Athens, city named for great Athena, honored
above all cities in the world!" (Soph. OC 107-8). Similarly, in the opening
speech of Euripides' Ion, Hermes refers to events that transpired in Athens and
mentions "the famous city of the Greeks called after Pallas of the Golden Spear"
(8-9); later reiterating a positive valuation of Athens, Hermes reports Apollo's
request to Creusa to take their newborn child "to the earth-born people of glo-
rious Athens" (29-30).I6 Moreover, the Chorus in Aeschylus' Persians refers
to Xerxes as "the peer of the gods, one whose race is sprung from gold (%p/ucrov6|j,o'u

15. This passage bears similarities to Aesch. Eum. 8.

16. As the play draws to a close, Athena announces herself as "the one who gave her name to
your land (emavujioi; ofi<; %6ovoc;)" (1555; my translation).
106 THE ANCIENT QUARREL

yevedi;)" (8o);17 the poet refers here to the hero Perseus, who was conceived
following Zeus's approach to Danae in a shower of gold.18 Strictly speaking,
the implied judgment of desert pertains not to the Persians as a group but to
an individual Persian. Nevertheless, Xerxes is no ordinary Persian but his
people's supreme leader and representative.
In the unsystematic usage of literary eponymy, there is wide variation in the
identity of primary and recipient entities, which include mortal individuals and
groups, natural inanimate entities like rivers, and human constructions like
temples. Although, with rare exceptions, the primary referent is singular, the
primary and secondary referents may be singular or plural. In addition, there
are no determinate criteria governing which types of entities can serve in each
capacity and which sorts may be paired with one another. The same type of entity,
for instance, a mortal individual or group of such individuals, can serve in dif-
ferent contexts in either a primary or a recipient capacity. Thus, groups are of-
ten named after individuals (the lonians after Ion, the Persians after Perseus,
and so on); a group may also be in the primary role, however, as when it gives its
name to a parcel of land (see Hdt. 4.204, where a village is named "Barca" after
the Barcaeans). While the literary tradition often uses eponymy to make identi-
fications of source, it chooses as source the element that appears salient in a given
context and is not concerned to achieve a scientific understanding of the na-
tures of the entities in question.' 9

1.3. Platonic Eponymy and Its Metaphysical Backdrop

The framework of eponymy is potentially useful to Plato because it parallels,
in general terms, the structure of his metaphysics; that is, it accommodates two
referents in a relation of primary to secondary. In order to make it philosophi-
cally viable, however, Plato must institute key revisions of the literary tradition's
approach. Since Plato's reconstitution of eponymy is necessitated by his meta-
physics, before addressing questions of semantics I must comment briefly on
his theory. 20
At the heart of the metaphysics of Plato's middle dialogues is a dichotomy
between reality (namely, the totality of Forms) and appearance. Forms, which
exist separately from the spatiotemporal realm (Phd. y8d, loob; Symp. znb;

17. Tr. Smyth (1922), slightly modified.

18. Regarding Pers. 146, see n. 8 above.
19. Greek authors' focus on identifications of source in their use of eponymy fits in, gener-
ally speaking, with attempts by poets like Homer and Hesiod to offer explanations of important
phenomena and states of affairs in terms of their origins. Julius Moravcsik labels this approach
the "productive model"; for discussion of this and other explanatory patterns in Greek thought,
see Moravcsik 1991.
20. For sustained discussion of a range of issues in Platonic metaphysics, see Ross 1953; Allen
1965!); Malcolm 1991; Moravcsik 1992.
 CONCEPTIONS OF APPROPRIATENESS 1O/

Rep. 5i6b), serve as "explainers" (aitiai) with regard to pertinent entities in the
realm of appearance. 21
In contrast to Parmenides, who viewed appearances as categorically fallacious,
for Plato they need deceive us only if we construe them as the sole existents.
While the Eleatic philosopher had distinguished between what is and what is
not, admitting no middle ground between these extremes, Platonic metaphysics
carves out a place for that which is between (metaxu) "what is absolutely and
what in no way is" Rep.
478d6-7).
Having acknowledged the existence of those intermediate entities that in his
view constitute the domain of appearance, Plato introduces the concept of "par-
ticipation" to connect them to what is real. By speaking of participation, Plato
wants to convey the idea that entities belonging to the relevant classes in the
spatiotemporal realm have the properties in question (e.g., justice and equal-
ity), but always with some qualification. In the Phaedo, he introduces participa-
tion by insisting that "if anything else is beautiful besides the Form of Beauty,
there is no ground on which it can be beautiful save by participating in that Form.
And the same principle applies in all other such cases. Do you agree to a ground
of this kind?" (e'i it ecruv ciA,X,o Kcdov 7tA,T)v amo TO KaXov, o\>5£ 81' EV aMi
KaAov eivca f) 81611 p,eTe%ei eKeivoit Toti KaXoar Kai TKXVTOI 8r| oihax; ^eya). ir\
TomSe aitta auyxcopeit;;, 10004-7). Here Plato provides as the ground (aitid)
of any other entity's being beautiful its participation in the Form of Beauty and
states that one can treat all other relevant cases on this model. Following remarks
on the Large and Small (looe—loib), Plato turns to arithmetic operations and
considers what the proper explanation would be of why the result is 2 when i
and i are added. He advocates resolution of this and related questions by appeal
to the general principle that there is no other way in which each thing comes to
be "than by sharing in that specific nature in which it participates" (i\ |0,etao%6v
Tf\q i5ia<; owiac; eKaoiot) ou av \iemorxr\, 10103-4). Although Plato uses the
unmodified participle yiyvop.evov ("comes to be," 03), context shows that the
appropriate qualification is assumed; that is, here—and more generally—he is
asking, not how something comes to be from nothing at all, but how each thing
comes to assume a certain range of attributes.
This reference to "a certain range of attributes" raises the issue of the scope
of the theory of Forms and hence of Platonic eponymy. The ultimate meta-
physical foundation of any argument that eponymy constituted a general theory
of predication would be the assumption of Forms for all common nouns and
adjectives. This stance is, however, problematic. In the Phaedo, Plato under-

21. On Forms' separateness, see also Prm. i2gdy—8 and ijob—d, where the term choris appears
on multiple occasions, and 12866—12931 (auto kath' hauto). Notably, Aristotle identifies separa-
tion as what distinguishes Platonic Forms from the objects of Socratic inquiry (EN 1096!}; Metaph.
gSyb, io86b). Key passages from the Phaedo involving the reality-appearance dichotomy and
Forms' role as explainers include 74d-e, ySd-e, and ygc-d.
108 THE ANCIENT QUARREL

scores particulars' inevitably deficient embodiment of those properties whose

instantiation ties them to Forms (see especially j^d—e). The concept of defi-
ciency is also prominent in the Symposium (2106-2123) and Republic (particu-
larly Books 5-7). In the Republic, moreover, Plato distinguishes explicitly
between judgments of objects like fingers, which may be made adequately via
perception (aisthesis), and assessments of properties like smallness that neces-
sitate the existence of Forms to which the intellect (noesis) may appeal (523-
524). Plato thus recognizes a class of properties in discussions of whose posses-
sion it is not fitting to speak of deficient or qualified embodiments, and hence
of Forms: Something either is, or is not, a finger. While one may need to adjust
one's vantage point (e.g., move closer to the object in question) in order to
make this determination, the perceptual viewpoint from which a firm judg-
ment can be made is in principle achievable. 22 Where qualified embodiments
are at issue, in contrast, there must be Forms to serve as the reference points
of reason's assessments.
Despite the presence of such textual evidence, some commentators, relying
heavily on Republic 59636-7, have concluded that Plato espoused a theory of
universals along the lines of Bertrand Russell and G. E. Moore.23 In this famous
passage from the dialogue's final book, Plato says that "we are accustomed to
posit a single Form corresponding to each multiplicity to which we apply the
same name"
Republic 5963 is the only place where
Plato makes a statement of this type. Although Annas (1981, 227) grants that "the
reference to 'usual way' and 'we 3re accustomed' suggests some background of
which we have no further notice," she endorses a broad interpretation of the
passage. In her view (228), this construction, which puts Plato's remark here at
odds with his restrictive statements about Forms elsewhere in the middle dia-
logues, is nevertheless "more in tune with the other oddities of Book 10." As
Moravcsik (1992, 66) has argued, however, Plato lias in view here his own cus-
tomary method, and reflection on that approach gives one good reason to think
that while the realm of Forms is expanded to include instances in which those
entities are idealized functional properties, he does not endorse a general
account. 24 Moravcsik concludes (1992, 68) that Plato's arguments in the Phaedo,
Symposium, and Republic yield Forms corresponding to "the sciences of ethics,
dialectic, mathematics, geometry, music, and medicine," along with "the pro-

22. Cf. the Parmenides' denial (ijoc-d) that one may speak of Forms in the case of entities
such as mud, hair, and dirt, which are clearly amenable to treatment along the lines of finger.
23. For a prominent example, see Ross 1953. Notably, Russell himself (1988, 91) construes
Forms along the lines of his universals (for his own stance, see Russell 1971, 1988). On the topic
of universals and particulars, see also Moore, Stout, and Hicks 1923.
24. According to Smith (1917), recourse to grammatical considerations in the analysis of
Rep. 59636—7, in particular of the relative clause (011^ TCCTJTOV 6vop.cc efti<j>epo|iev) in 37, supports
the view that Plato is not offering information here about the scope of the theory of Forms.
 CONCEPTIONS OF APPROPRIATENESS 109

duction and understanding of artifacts."25 If one accepts this view of the doctrine's
scope, which finds strong support in these three dialogues taken as a group, it
becomes clear that, far from constituting a general account of predication, the
semantic correlate of Plato's doctrine of Forms covers only a limited subset of
cases in which Moore and Russell invoke universals and instantiation.
I round out these brief remarks on Plato's metaphysics by stressing that the
concept of "nature," in whose treatment he diverges both from literature and
from Presocratic natural science, plays an essential role in his theory. What is
natural for Plato is a certain articulation of reality into objective and theoreti-
cally important unities, which are granted the status of Forms; the key is to iden-
tify elements whose interrelations yield that order and harmony which he con-
strues as paradigmatic (the most central cases of this being entities in the realm
of mathematics and of values). Articulations of the spatiotemporal domain, in
contrast, are natural only in a derivative sense, based on the relations its con
stituent entities bear to the Forms in question. The Phaedo offers direct linguistic
testament to the centrality of naturalness, as when Plato emphasizes that the op-
posite itself can never become opposite to itself, "neither that in us nor that in
nature" (ome TO ev f||aiv owe TO ev rr\ ^iJoei, 10^5). The opposite "in nature" is
none other than the Form itself.26 Elsewhere in that dialogue, he invokes "the true
earth" (f) yn CTUTTI), plus the superior quality of what issues from and belongs to it,
as an allegory for the realm of Forms and the qualitatively higher status of its con-
stituent entities (1093-1140; for occurrences of the terms phusis and phuein in this
section of the dialogue, see 10965,11033, d3~4> ma2; C4, £6, ii3di).
In the Symposium, moreover, Plato states that the lover who completes the
ascent will encounter something astonishingly beautiful in its nature (TV Goru^acrcov
TT)V (^wiv KaA,6v, 21064-5).27 The Republic too provides salient linguistic evi-
dence, as when Plato asserts that "it is in the nature of the genuine lover of learn-
ing to strive for access to reality. A person of this type does not remain intent on
the multiple individual sensibles that are the objects of belief but progresses,

25. The status of natural kinds is difficult to sort out. As previously observed (chap. 3, n. 43),
it is not evident whether the Phaedo acknowledges Forms for fire and snow. The Timaeus does
admit Forms in the case of some natural kinds. If one views the Phaedo's position as unclear,
then one's conclusion about when Plato definitely recognizes Forms of this type will hinge on
one's dating of the Timaeus.
26. For the phrase "in nature" (ev ii\ <|>vjaei) used with reference to the domain of Forms, see
also Rep. 59yb6, 59831. On "nature" in the Phaedo, cf. 10369-10^4, where Plato uses phuein twice
when speaking of necessary relations between Forms, the idea being that some Forms necessar-
ily have other Forms as attributes. Plato need not apply this line of reasoning directly to the rela-
tion between Oddness itself and its own participants in the spatiotemporal arena because the
relevant conclusions are obvious based on what he has already said about individual Forms, for
example, auto to ison ("the equal itself). What Plato instead has to argue for is that one ma
explain certain necessary relations between Forms themselves in an analogous manner.
27. This use of phusis contrasts sharply with Euripides' linkage of Polynices' name and nature
via the verb phuein (Phoen. 1493), which reflects that rival view of "nature" targeted in the Cratylus.
110 THE ANCIENT QUARREL

without becoming disheartened or desisting, until he gains access to that which

is truly the nature of each thing (npiv 00)101} 6 ecmv EKCIOTOU ifji; (jmoecoc;
m|/aa0ca)" (49038-^3). Subsequently, Plato maintains that those sketching the
constitution of the just city must look "toward what is by nature just (TO (|xu<je
5i.Kcaov) and beautiful and moderate and all such things" (501132-3); to phusei
dikaion serves here as a locution interchangeable with auto to dikaion. Elsewhere,
Plato speaks of the need for prospective philosophers to become skilled in cal-
culation (logistike); they must pursue it, not as dilettantes, but so as to catch a
glimpse "in thought of the nature of numbers (TTJ<; TCQV dpiGjirov ((moecQi;)" (52502-
3). In what follows, he refers to calculation as leading the soul upward and com-
pelling it to discourse about numbers themselves (jtepi ctmcov TCOV dpiGiicov)
(^d^-j); "numbers themselves" (amoi oi dpiGiiol) are of course among Pla-
tonic ousiai and are synonymous with "the nature of numbers" (f| TCOV dpiGfiCOV
(Ixuoi^). Such linguistic evidence documents Plato's correlation of phusis first
and foremost with immutable essences.
Plato's metaphysical picture raises a new and important semantic question:
what is the connection between the employment of a term (e.g., "equal") with
reference to a Form and its use as an ingredient in a series of complex predi-
cates (e.g., "equal in length," "equal to one person but not to another")? Putting
the question in a slightly different way, how is one to represent the semantics of
the Form-participant relation? For this purpose, neither synonymy nor hom-
onymy will do. Synonymy, according to which the term "equal," for example,
would apply in the same manner both to the Form and to pairs of sticks in the
spatiotemporal realm, is inadequate because it provides no way to represent the
derivative status of the second application of the term. Homonymy, in turn, would
involve equivocation and leave one with no way to represent any connection at
all between the two usages. A third device is thus required to show that the usages,
though not identical, are related and, moreover, that they are connected in such
a way that the second application is derived from the first. For this purpose, Plato
selects a device intermediate between synonymy and homonymy, namely,
eponymy. These distinctions, it should be emphasized, are ultimately conceptual
rather than terminological since, as the discussion of literary eponymy made clear,
the notions in question may be given linguistic expression in multiple ways.28
Literary and Platonic eponymy both involve two relata, with priority given to
that entity viewed as primary. Plato's conception of reality, however, leads him

28. Aristotle inherits the semantic relation of eponymy from Plato and modifies it to suit his
own philosophical purposes. Pros hen legesthai, or "focal meaning" (following Owen ig86b), which
functions as eponymy's successor, is viewed by Aristotle as accommodating a wider range of ties
between primary and secondary referents and as providing a better semantic correlate to his on-
tology. Regarding focal meaning, see EE 1236316-22; EN io96b26-2g; Metaph. 1003333^6,
1030329-30, io6ob3i-io6iaio. For a helpful recent discussion, one may consult Ward 1995. On
the question of whether Aristotle's account of the nutritive soul (DA 2.3) involves the type of pri-
ority at issue in focal meaning, see Ward 1996.
 CONCEPTIONS OF APPROPRIATENESS 111

to institute strict requirements on the identity of primary and secondary entities;

these requirements are different from those of the literary tradition and differ-
ent for entities in each of the two Platonic roles. The two types of entities must
be on separate planes of existence, and there is an absolute prohibition on those
in one class serving in the alternative capacity. Moreover, the primary entity must
always be singular, and those entities named after it must be at least potentially
a plurality. In a marked departure from the literary tradition, Plato specifies and
discusses the kinds of priority enjoyed by the primary entity, or Form—viz.,
ontological, natural, explanatory, and logical; its primacy rests above all on on-
tological grounds. These specifications, in turn, yield a set of conditions under
which one may properly apply that entity's onoma to a certain class of recipi-
ents. The framework of eponymy lends an optimal structure to Plato's reflec-
tions because it allows him to speak of naming a primary entity and of naming
derivatively other entities that share the nature of the former, but only partially.29
In its basic structure, Plato's approach most closely resembles those cases of lit-
erary eponymy in which a single individual gives his name to a group of indi-
viduals. However, this type of case, like the others, is ultimately unacceptable to
Plato as a model for philosophical usage because both the primary and recipi-
ent entities are native to or wrongly aligned with a single plane of existence,
namely, the empirical world; since the primary entity is misidentified, one can-
not assign it the requisite priority.30 In addition, the recipient entities in ques-
tion are groups of mortals, specifically, peoples, hence merely a subset of par-
ticulars in Plato's sense.
In contrast to literary practice, Platonic eponymy rests on an explicit and
fundamental ontological asymmetry between the two relata. Phaedo 10363-7
illustrates Plato's view of appropriateness when Forms themselves are at issue:
"It is always appropriate for the Form itself to have its own onoma applied to it,
without regard to time. It must surely always be fitting for Oddness to be in pos-
session of the onoma that we are now employing to designate it" (amo TO El8o<
d^iouaSai TOU amo'O 6v6|iaTOc; ei<; TOV del %povov. . . . TO ... rcepmov de
Tiox) 8ei Tomot) TOU ovoiamoc; Tuy%dveiv onep vw A£yop.ev). It is always appropri
ate to call the Form, to peritton, or "the odd," "odd" in an unqualified way. To
explain what undergirds this judgment from a metaphysical perspective, one must

29. For the latter application of terms as "derivative," cf. Code rg86, 427.
30. A distinction between nontechnical and philosophical usage is necessary here. Plato is
not saying that the city of Athens, for example, really gets that name based on an ontological re-
lation it bears to a Form rather than, say, from the goddess Athena. He is simply not interested in
providing revised standards for or eliminating such nonphilosophical uses of eponymy. Support
for this view is provided by the fact that Plato never criticizes nontechnical usage and by his
employment of it even after his introduction of a revised version of eponymy (as, e.g., at Leg.
626d3~5, where he presents the familiar derivation of "Athens" from "Athena"). Plato's efforts at
the revision of conventional usage are directed solely toward that range of cases in which a term's
primary referent must be separate from the spatiotemporal realm; for those cases, no literary pre-
cedent may serve directly as a model.
112 THE ANCIENT QUARREL

invoke the characteristic, being F in an unqualified way, that distinguishes a given

Form from its own participants. It is crucial to note, moreover, that Forms could
not be characterized in this manner unless they possessed those features (e.g.,
being unchangeable and imperishable) that, taken together, distinguish elements
of reality from those of appearance.31 While eponymy, in the strictest sense, covers
the derivative assignment, the two assessments of correctness are not detachable
in Plato, as they are in literature. As previously noted, when literary authors wis
to address the fitness of a primary referent's appellation, as in the cases of Ion
and Athena, they invoke a different notion, namely, etymology. The two analy-
ses, moreover, are not correlated in a dependence relation such that the fitness
of the eponymous derivation is made to hinge on the outcome of the etymol-
ogy: one need not, for example, hold up one's judgment about whether the
lonians are named properly after Ion until one investigates separately whether
the name "Ion" is fittingly assigned to its bearer (i.e., whether the name's descriptive
content discloses the nature of its referent). In the case of Plato, in contrast, it is
because particular entities in our world sustain the requisite ontological depen-
dence on entities whose own onomata single them out in a unique manner that
the secondary designations at issue in eponymy are possible. For this reason, one
may view the primary designation as pertinent to, or encompassed by, the frame-
work of eponymy on Plato's construction.
Plato first discusses the naming of particulars in the Affinity Argument, when
he refers to each class of individual sensibles as homonumon with the relevant
Form (Phd. ySez). As has been recognized, though Plato here uses homonumos
rather than eponumos or a related term (i.e., eponumia or eponomazein), he has
the phenomenon of eponymy in view.32 Notably, the literary tradition also uses
homonumos to single out the phenomenon of eponymy, as when Pindar men-
tions that Strepsiades the Isthmian victor got his name from his maternal uncle
(Isthm. 7.24). In such cases, whether or not individuals are homonumoi is a matter

31. Regarding Forms' characteristics, see Moravcsik 1992, 69—81.1 offer a sketch of Moravcsik's
framework below.
32. See, e.g., Gallop ([1975] 1988, ad loc.) and the contention of Cherniss (1962, 178 n. 10
that the particular "is 6|j.cov\)(j,ov tcp e'i8ei . . . because it has its name and nature derivatively from
the idea." For relevant instances of homonumos in later dialogues, see Tim. 5234-5 (TO OIKBVUUOV
6|o,oiov TE eiceivcp) and Prm. i^dz-y (id nap' riulv taCta 6(i(6v\)|j,a ovca EKeivon;). Cf. Aristotle's
use of this adjective in describing Plato's theory at Metaph. g8-/b-/—io (adopting Jaeger's text); his
comments reaffirm that, according to Plato, it is due to their participation in Forms that aistheta
share their appellations, are in fact named after them. (Strictly speaking, all entities called aistheta
by Plato are not perceptible since the relevant class includes prevailing conventions, notably, about
values. For present purposes, I follow Plato's own usage, with the understanding that the class in
question includes entities that are not directly accessible by way of the senses.)
Conversely, it is also worth observing that Plato uses eponumos, eponumia, and eponomazein
in multiple contexts. With regard to nontechnical usage, see n. 30 above. In addition, in numer-
ous cases Plato follows the literary tradition in employing this terminology in connection with
etymologies: see, e.g., Cra. 395b6, 397d4~5, 3g8ci, 4oob2,40^3,40335,40^2,40636,40907,4120
 CONCEPTIONS OF APPROPRIATENESS 113

of choice or agency. While Plato accords with the literary tradition at ySez in
his use of terminology, he differs fundamentally from it in the reason given for
the designation, that is, in his insistence that the selection of those entities des-
ignated as homonuma with a given Form never depends on individual choice,
whether mortal or divine, but solely on whether the secondary referents stand in
the ontologically proper relation to that Form. I therefore cannot agree with
Owen (19863, 171), who criticizes Cherniss's claim (1962, 178 n. 102) that the
particular "is 6tic6vuu,ov TCQ eiSei. . . because it has its name and nature deriva-
tively from the idea" by observing that homonumos may be used —as at Republic
33ob2~3 (6 nannoc, TE KOI 6|a,a>vuu,o<;) — "of an ancestor from whom the name is
derived." Even where this happens in literary and Platonic instances of ordinary
eponymy, the relevant nonsemantic dependency is never in doubt; in such cases,
the term's grammatical function is of secondary importance, and homonumos
must be interpreted in light of this dependency. Moreover, appeals to ordinary
practice cannot serve, in themselves, as evidence against a particular view of
Plato's use of the relevant terminology and concept of eponymy in connection
with the theory of Forms.
In all cases of eponymy, both literary and Platonic, the semantic relation
between the primary and secondary referents presupposes, and is based on, a
nonsemantic, independently specifiable link between them. In literary eponymy,
the nature of that association varies widely and depends solely on the context. It
may be familial in nature; for instance, male progeny are often named for their
grandfathers or other male relatives (see, e.g., Pind. Isth. 7.24, Ol. 9.63-64; Hdt.
6.131.1 [cf. 5.69.1 , 8.136.1). Elsewhere, the connection between entities is spa-
tial or geographical, as when a town lends its name to the gulf on which it is
located (Hdt. 7.121.1) and a kind of horse gets its onoma from that of the plain
which is its geographical source (Hdt. 7.40.3)." Literary authors might also focus
on the commemoration of an ordeal endured (Aesch. PV 840-41; Eur. IT 1454).
Sometimes an assignment is honorific, as when a treasure is named after its donor
(Hdt. 1.14.3) and a body of land is named for an iatromantis who had once served
as its benefactor (Aesch. Supp. 260-61). The foregoing list of categories may
generate an appearance of well-ordered diversity in the literary evidence. In point
of fact, however, distinctions between categories are often not clear-cut, as is
evident from the many instances in which multiple factors are salient. For example,
protective and honorific associations may intersect, as in the complex relation-

41362, 4i5b5, d3, 416011, d8, 41709; Phdr. 23803, 24408; Leg. Szibg. He employs it, moreover, in
connection with diairesis (sec, e.g., Soph. 22503, d4, 22gd6, 26782; Pol. 2&3di) and with regard to
the virtue-virtues relationship (Leg. 96305^7, with eponomasamen at d6). Such uses are not rele-
vant to the present argument, which concentrates on the semantics of the Form-participant rela-
tion, where the term's primary referent exists apart from the spatiotemporal world, and all of its
secondary referents are native to that realm.
33. For other relevant cases, see Hdt. 4.52.1, 4.184.3-4.
 THE
114 ANCIENT QUARREL

ship between Athens and Athena (Soph. OC 107-8; Eur. Ion 8-9, 29-30,1555-
56). Elsewhere, political and honorific associations are interwoven (Horn. 11.
20.230; Aesch. Supp. 250-53), as are geographical and commemorative ties (II.
4.474-77). In one noteworthy case, Archias reports that his father was named
"Samius" to commemorate the heroic death of Archias' grandfather in Samos
(Hdt. 3.55.2); here familial, honorific, and geographical associations converge.
In sharp contrast to the literary tradition, Plato uses one basic nonsemantic
relationship, that of participation, to undergird eponymy. Since that ontologi-
cal relation provides the foundation for eponymy, it is not surprising that he takes
up the eponymy relation most directly in connection with his treatment of par-
ticipation. Plato moves from the participation relation to that of eponymy with
the agreement of those present that the Forms exist and that "the other things,
participating in them, are named after the Forms themselves" (tomcov tfiMxx
|a,eTdA,ci|ipdvovTa amrov TOTJTCOV ii\v e7i(ovu(j,iav 'io%eiv, Phd. loaba-j). This
passage highlights the fact that it is due to their participation in Forms that indi-
vidual sensibles are named after them (note the participle's causal force).34 Plato
returns to these issues a little later. Speaking generally, he sums up the contrast
as one between things that have opposites, which are named after those oppo-
sites, and "those opposites themselves from whose inherence in them the things
named receive their appellations"; that is, Plato emphasizes that while the earlier
discussion was about TCOV exovicov id evavcia . .. ejiovond^ovTec; amd T^
eKeivcov eTtcovuuia, now it concerns eKeivcov oancov cov evovicov e%ei tf)v
ETicovuniav id 6vona^6(xeva (io3b6-ci). Here Plato raises the same metaphysical
and linguistic questions as he had in the earlier passage; b8 contains language
of inherence rather than that of participation because there he considers the
Form-particular relation from the point of view of Forms rather than that of
individual sensibles.
The twin issues of self-attribution (or self-exemplification) and the Third Man
Argument (TMA), along with the immense controversy surrounding them, fall
largely outside the scope of the current project. 35 My comments on these topics
will be, therefore, quite brief and selective.36 Some, including Allen (19653,45-
47) and Bestor (1978), have claimed that one may invoke eponymy to block the
regress at issue in the TMA. On this interpretation, onomata when applied to

34. For this same use of the participle, see Prm. 13065-6: Soicei aoi, w<; ((rqi;, elvai e't5r| area,
rov xd8e id 6iU,a \i£.ia\a\i$avov%a rac; ejtcflvuplat; amcov io~xeiv. Cf. i33ch-2 (on the assump-
tion, which I endorse, that the passage does not deal with immanent characteristics): cov iinei<;
|i£T.e%ovTe<; elvcu eicacrta ETtovoiia^oiieSa.
35. For the term "self-exemplification" regarded as preferable to "self-predication," see
Malcolm (1991, i [with n. 4]), who indicates that in favoring it he follows Moravcsik (1976, 14).
For the term "self-attribution," see Moravcsik 1992, 138.
36. For a sample of views on self-exemplification and the TMA, see Vlastos 19653,1965^ 1969,
1981; Malcolm 1981,1985,1991; Moravcsik 1963,1992; Nehamas 1979; Bestor 1978; Meinwald 1991.
For additional references, see the bibliography of Malcolm 1991.
 CONCEPTIONS OF APPROPRIATENESS 115

Forms serve as proper names, such that "to say that F-ness is F is to state an iden-
tity" (Allen 19653,46).37 While the precise details of Plato's stance are not trans-
parent, what is clear, I believe, is that Forms' onomata cannot operate in this
way. The central reason for the inadequacy of this view is that, according to Plato,
terms such as kalon have descriptive force when applied, not only to Forms'
participants, but also to Forms themselves. Thus, in the Symposium, Plato depicts
the Form of Beauty (auto to kalon) as "astonishingly beautiful in its nature"
(21064-5; cf- 2ub2, where Plato contrasts it with "all the other beautiful things,"
TO ... ciA.A.a TtdvTd KCtka). In the same vein, in the Phaedo, he claims that "if
anything else is beautiful besides the Form of Beauty (TtMrv curro TO KOtA,6v), th
only way in which it can be beautiful is by participating in that Form" (10004-6).
This feature is far from being a minor element of Plato's theory. As Moravcsik
(1992, 69-81) has argued, the middle dialogues invest Forms with three types of
characteristics: those, for instance, imperishability, that they share with other
entities like souls; those distinguishing Forms as a class from other kinds of en-
tities, for example, being such that other things can partake of them; and, finally,
that characteristic which marks off individual Forms from one another, namely,
being unqualifiedly what their participants are only with restrictions. Although,
as Moravcsik observes, Plato takes the pressure off this final characteristic in later
dialogues' handling of individuation (1992, 138, 169), in the middle period its
role is pivotal. Also central is Forms' explanatory role vis-a-vis classes of entities
in the spatiotemporal realm: they are to account specifically for these entities'
qualified embodiment of properties such as beauty and justice (see, e.g., Phd.
looc, e), and it is quite difficult to see how they could accomplish this unless
Plato held the view of Forms' natures reflected in the above comments from the
Phaedo (10004-6) and Symposium (21064-5, 211b2).
Another point of contention is whether the Form is or has the nature in ques-
tion (or, perhaps, at once both is and has it). Plato uses the terms phusis and
phuein in two central ways in his discussions of reality. He employs phusis, first,
to denote the domain of Forms in its entirety (Phd. iO3b5; Rep. 59yb6, 59831).
Plato's usage reflects his view that this sphere, rather than the familiar world whose
operations were often the focus of earlier philosophers' attention, constitutes
nsture proper. Plato employs the term phusis, in addition, with reference to those
individual Forms that comprise the "world" viewed by him as truly natural. A
central opponent here is the view, prominent in the literary tradition, that na-
tures are associated first and foremost with individusls, that is, with certain con-
stituents of, or entities closely tied to, the spatiotemporal world. Thus, he says
that the Form of Beauty is "astonishingly beautiful in or by its very nature"
(9a\)|iaoT6v Trvv 0iJOiv KCtAov, Symp. 21064-5) anc' that the ideal city's founders
must concentrate on "what is bv nature iust and beautiful and moderate and all
such things" (

37. For an argument that the regress cannot be blocked via this route, see Malcolm 1981, 287.
116 THE ANCIENT QUARREL

Rep. 5oib2-3). It is hard to determine from such remarks whether the Form is
or has the nature at issue (or both). Two passages from the Republic might be
interpreted to indicate a commitment to the view that, at the very least, the Form
is a nature: that in which he speaks of the philosopher's access to what "is truly
the nature of each thing" (49038-^3), and that where he insists that philosophers-
to-be pursue calculation in order to glimpse in thought "the nature of numbers"
, an expression synonymous with "numbers them-
selves" (amoi oi dpi9jioi, dj-y). These remarks, however, while suggestive, do
not permit a firm resolution of the controversy about the phuseis of individual
Forms.
I therefore conclude that the terms in question (e.g., kalon and dikaion)
apply in some way that is descriptive to the Forms, but that the precise charac-
ter of their descriptive force cannot be specified. When speaking of descrip-
tion, it is crucial to distinguish between descriptive content and descriptive
force. The claim being attributed to Plato in the present investigation is not
that analyses of individual terms' descriptive content disclose their referents'
natures. As I argued in chapter 3, this approach does not fare well in the Cratylus.
The point is rather that the term kalon does not merely pick the Form out, as
my name does me, but gives one valuable guidance about what it is to be that
entity.
We are now in a position, finally, to complete our answer to the question
of whether Plato embraces nature or convention, or some combination of the
two, as pertinent to judgments of fitness. As we saw in the previous chapter,
the Cratylus endorses convention as the basis of correctness when terms' con-
stitution is at issue. Yet, as noted there, this conclusion does not exhaust the
scope of what is presupposed by or involved in the making of such assessments.
Above all, one must have the right conception of reality, that is, the correct
view of what is or has a nature (phusis) in the primary sense. As the Cratylus
makes clear, Plato does not view natures as belonging in the first instance to
mortal or divine individuals (e.g., Atreus or Apollo). While the end of the dia-
logue points one in the direction of Plato's favored conception (438-440), the
elaboration of this stance is reserved for the Phaedo, Symposium, and Repub-
lic. In particular, Plato's use of the term phusis in all three dialogues fore-
grounds the fact that his own conception of "nature," which underlies correct
assessments of terms' application, differs radically from that view challenged
in the Cratylus. One may note, finally, a refinement of the Cratylus' picture
with respect to names. In the Phaedo, as in the Cratylus, Plato adheres to the
view that the constitution ofonomata, even those such as kalon whose primary
referents are Forms, is appropriate based on convention. Once the phrase auto
to kalon ("the beautiful itself) is chosen to designate the Form, however, it is
not simply a matter of convention that its participants, which are the term's
secondary referents, are called ta kala. Combining the Cratylus and Phaedo,
one thus ends up with a framework consisting of the following three variables:
the identification of those entities that have the status of Forms, which is, ac-
 CONCEPTIONS OF APPROPRIATENESS ll/

cording to Plato, in no way governed by convention; judgments of fitness with

regard to the constitution of individual terms, which are the province of ethos
and suntheke; and the designation of a Form's participants once that Form is
labeled in a particular way, which is not merely a matter of convention since
the linguistic tie forged by the assignment is intended to reflect a metaphysi-
cal connection between the relata.

1.4. Conclusion
In the Phaedo, as we have seen, eponymy is important because it provides a frame-
work within which Plato can present his theory of terms' proper employment,
which is based on his metaphysics and involves fundamental revision of con-
ventional usage of the onomata in question.38 That eponymy remains an impor-
tant notion for Plato is evidenced by its presence, not only in the Phaedo, but
also in later dialogues, namely, the Parmenides and Timaeus, when he wishes to
characterize the different ways in which terms apply to Forms and participants.
Indeed, the issue of how one may characterize the semantic relation between,
for example, "two" as the name of a number and "two" as it appears as an ingre-
dient in complex predicates has continued to occupy philosophers, notable
among them Gottlob Frege.39
Eponymy as discussed by Plato applies solely to relations involving Forms
and their participants in the spatiotemporal realm. When Plato attempts in later
dialogues to address objections to his middle-period metaphysics, prominent
among them the Third Man Argument of the Parmenides, he shifts the focus of
his technical discussions to mutual connections between Forms.40 To describe

38. Regarding conventional usage, human beings are seriously deceived if, when hearing the
term kalon, they are unable to distinguish between qualified and unqualified applications of it,
an incapacity exhibited strikingly by "the lovers of sights and sounds" of Republic 5. On this con-
fusion with regard to kalon, cf. Hippias' response to Socrates' query (it EOTI TOOTO TO KaX,6v;)
and the ensuing discussion between them in the Hippias Major (i8yd ff.). On the question of the
dialogue's authenticity, see the sustained defense by Woodruff (19823) and the case for non-Platonic
authorship in Kahn 1985 (cf. 1996, 182).
39. On the secondary use of terms in Frege, see, e.g., Frege 1950, 59. The more general con-
cern of Frege (1950) is to identify numbers as objects so that one can talk about them without
having to add any common noun (aside from "number") to supply a principle of individuation.
Hence, notwithstanding the existence of important differences in their views, both Plato and Frege
clearly distinguish primary and secondary uses of a key range of terms.
40. Plato sometimes wishes to emphasize the sharp difference between Forms and particu-
lars, at other times their relatedness. Perhaps he saw that his theory required both perspectives or
dimensions but could not determine how to retain them adequately within the confines of a single
account. What is clear is that Plato develops the theory of Forms in later dialogues so as to stress
those entities' interconnections. Whether this was because he thought that the original theory
was correct but could not sec how to develop a compelling argument to buttress his intuition on
this score, or because he worried that his argument as it stood was vulnerable to the TMA, is dif-
ficult if not impossible to determine.
118 THE ANCIENT QUARREL

these ties he introduces another notion of partaking. The semantics of that rela-
tion falls outside the scope of the present inquiry. 41
What I hope to have shown here is that although Plato rejects decisively lit
erary antecedents which he views as philosophically unpromising, he does not
hesitate to embrace notions that do have potential. As I have argued, in his phi-
losophy of language Plato in fact does both. In the Cratylus, he criticizes the
literary tradition's use of etymology. Moreover, Plato's disbelief in the philosophi-
cal value of an attempt at its reconstruction is evident both in the fundamental
nature of his challenge to literary practice and in the fact that what hints he does
provide at the close of the dialogue of the form his own theory will assume point
in quite a different direction. The Cratylus engagement with the literary tradi-
tion thus terminates in Plato's rejection of one manifestation of its handling of
fitness. In the Phaedo, in contrast, where Plato seeks a framework in which to
express his own insights about orthotes onomaton, he singles out a second key
device prominent in literature, namely, eponymy, and adapts it to suit his meta-
physics centering on Forms and participants.42 Eponymy is a central semantic
notion for Plato; literary uses of it, however, antedate Plato specifically and the
philosophical tradition generally. While in many cases we may understand Pla-
tonic doctrines against the backdrop of earlier philosophical and sophistic
inquiries, we can best appreciate what is distinctive about Platonic eponymy by
reflecting on its literary antecedent.
In the remainder of this chapter, I turn to Republic 5, where, for a third time,
Plato treats the issue of appropriateness against a literary backdrop. While Re-
public 5 is closely linked for this reason both to the Cratylus and to the Phaedo,
it has special ties to the latter insofar as here Plato once again gives serious weight
to his literary predecessors in the context of formulating his own position. In this
case, as we will see, Plato's theory of the proper allocation of sociopolitical and
familial roles leads him to assign an important philosophical position to a
revamped construction of yet another linguistic device prominent in literature.

41. Plato does not use eponymy to provide the semantics of relations between abstract enti-
ties. One who wished to claim that the ingredients of this extension are present in his writings
might focus on considerations such as the following: Plato's introduction of a notion of partaking
to cover links between abstract entities (see Soph. 2jid ff); comments like those at Phd. rooc/f-y
(quoted earlier, p. 107), where the words ti allo (04) arguably encompass not only ta kala but also
any relevant Forms besides auto to kalon that are suitably related to it; and Plato's use of
eponomazein in the Laws in a discussion of the connection between virtue and individual virtues
(963c5-d7).
42. I am in accord with the claim of White (1976, i^r) that the Cratylus does not have an iso-
lated position in the Platonic corpus. White sharply deemphasizes the existence of ties between the
Cratylus and Phaedo, dating the Cratylus after the Phaedrus and linking it closely to that dialogue
on thematic grounds ^976, chaps. 3, 5—6). While White grants that stylistic considerations favor an
earlier date, he contends that evidence regarding content points solidly in the other direction (148
n. r). As the foregoing investigation makes clear, in my view there are much stronger thematic grounds
for favoring close ties between the Cratylus and Phaedo than White acknowledges.
 CONCEPTIONS OF APPROPRIATENESS 119

2. Functional Terms: Literary Usage

and Plato's Revision Thereof

2.1. Introduction
In his discussion of the procreation and rearing of children in Republic 5, Plato
asserts that

a man will call all children born in the tenth or seventh month after he became
a bridegroom his sons (if they are male) or daughters (if they are female), and all
of them will call him father.43 He will call their children his grandchildren, and
they, in turn, will call the members of the group to which he belongs grandfa-
thers and grandmothers. Moreover, they will call those offspring born at the time
when their mothers and fathers were producing children their sisters and brothers.
The result will be that of which we spoke, namely, that the relevant groups avoid
sexual relations with one another. (46^2-62)

The following discussion of Plato's view of the family sets it against the back-
drop of literary reflections for the purpose of illustrating both where he aligns
himself with them and where the realization of his philosophical purposes ne-
cessitates his breaking new ground. In contrast to the discussions of etymology
and eponymy— which gave attention to Homer, Hesiod, Pindar, the three tra-
gedians, and Herodotus — here I focus exclusively on tragedy.44

2.2. Familial Roles and Responsibilities:

Literary Treatments of Appropriateness
In addition to raising questions of appropriateness in connection with etymol-
ogy and eponymy, the literary tradition does so with respect to functional terms
like "mother" and "son." In the case of what I am calling "functional terms,"

43. Regarding gestation periods, cf. Arist. GA yyzbS-io; HA 584336^15. On the numerologi-
cal ideas about pregnancy and childbirth expressed in Greek medical writings, see Lloyd 1987,
259-64.
44. For discussion of this difficult section of Republic 5, see, e.g., Adam 1902, vol. i, 292-315;
Grube 1927; Cornford 1941, 155-68; Ranldn 1965; Bloom 1991, 384-88; Halliwell 19933, 155-82.
For consideration of changes in Plato's recommendations concerning marriage between the
Republic and Laws, see Fortenbaugh 1975.
120 THE ANCIENT QUARREL

the primary meaning of an onoma derives from what its denotation does (thus,
e.g., the central meaning of the term "son" may be traced to the filial duties
associated with its referent). 45 In the context of tragedy, where such onomata
often have normative force, the existence of biological connections is not suffi-
cient or even required for terms' proper employment. As the following discus-
sion makes clear, one considers instead whether an individual performs the duties
implied by the term in question.
On several occasions, characters in tragedy question the appropriateness of
calling certain individuals "mother" and "father." Speculating about the source
of a lock of hair found at Agamemnon's tomb, Electra says: "Nor yet in truth
did she shear it from her head —she the murderess, my own mother, who toward
her children has taken to herself a godless spirit ill-according with the name
of mother" (
, Aesch. Cho. 189-91).46 Having
admitted the dubious appropriateness of using the term "mother" with refer-
ence to Clytemnestra — in light of her murder of Agamemnon and affair with
Aegisthus —Electra offers as possible alternatives "this woman who sleeps with
him" Soph. El. 274) and "mistress" ( 597).
Subsequently, when addressing what she takes to be Orestes' ashes, Electra drives
home the point that Clytemnestra's attitude and conduct entail her forfeiture of
the consideration owed one who does in fact merit the appellation (1145-56).47
Finally, Electra insists to Orestes himself that "she is called my mother—but
like a mother in nothing" (inrrrip KaXeuar uT|tpi 8' o\)8ev e^iooi, 1194); here
the discrepancy between appellation, on the one hand, and conduct or attitude,
on the other, is most explicitly formulated. 48
Deianira provides another target for the same charges. Mincing no words,
Hyllus maintains that Deianira's murderous conduct toward Heracles has destroyed
her entitlement to the respect and consideration ordinarily due one's mother
(Soph. Track. 807-12). In remarks calling to mind those of Electra, Hyllus
exclaims: "Why should she maintain the pointless dignity of the name of mother
when she acts in no way like a mother?"
, 817-18). Seconding Hyllus'
assessment, Heracles pleads with his son to deliver Deianira to him for retribu-

45. The purpose of this brief description is not to single out functional terms uniquely but
rather to set them apart from cases (e.g., those of rocks, stars, and triangles) in which terms' refer-
ents have functions without the roles in question being by definition attached to them.
46. Tr. Smyth (1926), slightly modified. Lattimore (Lattimore, Benardete, and Grene rgj^)
translates as follows: "She never could have cut it, she who murdered him and is my mother, but
no mother in her heart which has assumed God's hate and hates her children." I prefer Smyth's
rendering of these lines since it brings out the pivotal linguistic dimension of Electra's criticism.
47. See especially Electra's reference to her as "mother, no mother" (ufVcr|p ri|U.T|TCflp, 1154).
48. Euripides, in turn, makes Clytemnestra the target of an implied judgment of non-desert
with respect to her status as "mother" (see Or. 557-59, 585-86).
 CONCEPTIONS OF APPROPRIATENESS 121

tion and provides an alternative description of her (1064-67). Taken in sum, the
foregoing remarks about Clytemnestra and Deianira indicate that being someone's
mother in the biological sense is not sufficient to generate an entitlement to be
addressed by the appellation in question; rather, one's character and actions
provide the definitive criterion in light of which judgments of fitness are made.
Admetus, in turn, reproaches both his parents at once for their failure to of-
fer themselves up to die in his place. Since it is Alcestis who volunteers to per-
form this fatal act of substitution, Admetus concludes that he should be called
instead her child and support in old age (jr\poip6^oc,) (Eur. Ale. 636-49, 666-
68).49 Admetus indicates in no uncertain terms his disposition on the matter of
desert and at least voices reservations about his actual parentage; his remarks
suggest that the two planes are closely related and that it is his dissatisfaction
with the conduct of his "parents" that leads to this two-pronged condemnation
of them. Admetus' expressed doubts about his parentage do not reflect genuine
concerns on the biological plane; rather, his substantive complaints all involve
his parents' failure to meet what he views as their obligations toward him as their
son.50 The core of Admetus' reproach is that if his mother and father truly de-
served their respective functional appellations, or collectively the title of "par-
ents," one of them would choose to die in his stead; since his wife, Alcestis, makes
the sacrifice that Admetus believes to be their responsibility, he concludes that
it is she —rather than they—to whom the appellation "parent" is appropriate. 51
Admetus retaliates in kind, contending that while he has fulfilled his own filial
duties up to that point, he will do so no longer (658-66). In sum, then, in Admetus'
view his parents' failure to act in what he construes as the appropriate manner
toward him qua son deprives them categorically of the entitlement to be called
"parents" and of the right to expect his fulfillment of any of those responsibili-
ties previously linked with his role as "son." This case is particularly noteworthy
for its introduction of the idea that one who has a nonbiological kinship rela-
tion to the speaker may nevertheless be entitled to an appellation denoting bio-
logical ties based, once again, on considerations of attitude and behavior.
Elsewhere, discussion centers on the terms "daughter" and "son." Aeschylus
criticizes Helen via a judgment of appropriateness based on an etymological

49. The term yt|pOTp6<|>0(; appears in line 668. Notably, Admetus deliberately avoids as far as
possible using the term "mother" (see 637-39, 661); where it does appear (638), it is in a phrase (i\
KEKA,r|nevr| |rr|T,r|p) that includes a qualification.
50. For discussion of this issue, see Griffith 1978. In all literary cases discussed in this section
of the chapter, the speaker believes that his or her judgments and demands are reasonable, but
clearly distinctions can be drawn between various cases in this respect.
51. The fact of their advanced age is pertinent here (see 634-35, 648-50, 669-72); that is, it is
not clear that Admetus is claiming that his parents, at any age, must offer their lives up to save
his. What is indisputable, however, is Admetus' insistence that his aged parents' refusal to act as
he thinks obligatory deprives them henceforth of the right to receive the filial respect due one's
parents, or indeed to be called "parents" at all.
122 THE ANCIENT QUARREL

analysis of her name (Ag. 681-92). He does not raise the issue of paternity, being
content to use etymology to get his point across; in fact, far from questioning
Helen's status as Zeus's daughter, one might view the Chorus as alluding to his
possible role as namer of his progeny. Euripides takes a different tack when he
has Andromache question the appropriateness of calling Helen the "daughter"
of Zeus (Tro. 766-71). With regard to the term "son," Heracles informs Hyllus
that he has a duty to ease his father's plight by building a pyre and burning the
body, lest he forfeit his entitlement to be called the "son" of Heracles (Soph.
Track. 1199-1205). Moreover, acknowledging a possible disparity between onoma
and entitlement, Polynices tells Oedipus that "I'm your son, or, if not, . . .at least
I'm called (Ka\o\>[i£.vocs) your son" (Soph. OC 1323-24); also relevant here is
Oedipus' remark that Antigone and Ismene "have saved me, they are my sup-
port, and are not girls, but men, in faithfulness. As for you two, you are no sons
of mine!"
1367-69).52
While biological mothers' entitlement to the appellation of "mother" may be
called sharply into question, the process can also work in reverse; that is, an indi-
vidual's conduct may provide an otherwise nonexistent justification for calling her
"mother." In Euripides' Ion, the priestess at Delphi is said to have raised Ion from
infancy: "And so she took the child and reared him, not knowing who his mother
was, or that Apollo was his father; while the child has never known his parents"

, 49-51). Ion greets her with

Hoi |J,fJT£p, OTJ TEKOWCI nep, at which the priestess expresses her pleasure (1324-
25). Moreover, she bids him good-bye as follows: "And now farewell. I kiss you as
my son" (iced %aip'' 'ioov yap a' ax; TEKOW' dcmciCpum,, 1363). The participle in
line 1324 has clear concessive force (oij TEKO'Ocd Ttep); hence, the import of Ion's
remark is that although the priestess is not his biological parent, she deserves to be
greeted as "mother" based on her treatment of him. In the priestess's parting remark
(1363), in turn, hos combines with ison to suggest, not that in her view their relation-
ship loosely approximates that of mother and son while necessarily falling short,
but rather that in all essentials the connection between them has achieved that
status. While both Ion and the priestess readily acknowledge the absence of a biolog-
ical tie, their remarks show that the priestess's attitude and conduct generate a justi-
fication for her being called by the appellation of "mother." This case shows that,
far from being sufficient to ground judgments of fitness involving functional terms,
neither biological nor other kinship ties between "relatives" are even required.53

52. The former translation is mine, the latter that of Fitzgerald (1941).
53. With respect to judgments of appropriateness involving the term "father," Ion claims that
Apollo's attitude and behavior toward him justify his use of that onoma with reference to the god
(136-40). Notably, Ion offers this praise without realizing that Apollo actually is his father (109—
n). In fact, we know already from Hermes that Apollo's intention is to advance Ion's cause while
concealing his own paternity (69-73).
 CONCEPTIONS OF APPROPRIATENESS 123

2.3. Plato's Revamping of Literary Practice

in Light of His Conception of the Ideal Polls
As previously noted, early in the Cratylus Plato discusses one functional term,
"king" (pcraiAeik;, 393-394), only to refrain from treating onomata of this type
in the long inquiry that follows. In the Republic's handling of appropriateness,
in contrast, this type of term takes center stage. Plato, like the poets, distinguishes
sharply between the descriptive and prescriptive levels in the application of func-
tional terms. He takes quite seriously the prescriptive or normative use of these
terms in tragedy and incorporates the idea, emphasized repeatedly there, that
biological connections are not ultimately decisive when the goal is to determine
who may properly be said to be mutually "related" as, for example, mother and
daughter. 54 Most specifically, he accepts the idea that biological ties are neither
a sufficient nor even a necessary condition for the ascription of kinship ties in
the case of those guardians and their progeny who will populate the ideal state.
Plato is not content merely to specify that children born to guardians will be
taken over by officials chosen for this task (460^7-8), since this arrangement is
compatible, at least in principle, with parents' still having some relationship with
progeny they know to be their own. Plato insists that this recognition itself must
not occur, prescribing that "no parent will know his own offspring, nor will any
child know its parent" (
yovea, 45yd2-3). He adds that when nursing mothers are brought to the creche
in which their infants have been placed, the presiding officials "will go to great
lengths to ensure that they do not learn who their own children are"
46oc9-di). As Plato
stresses at 46id-e, kinship distinctions in his ideal state are to be based, not on
biological ties, but instead on those conventions established by its founders.' 5
Like the tragedians, moreover, Plato distinguishes explicitly between being
someone's kin in name only and performing the actions characteristic of one
with a given role:

As legislator, will you stipulate merely that they use the relevant kinship terms,
or will you insist that they also do all the things that are implied by those terms?
For example, must they show to those designated as their fathers the same re-
spect, solicitude, and obedience currently due to parents? Won't they be held in
low esteem by gods and humans if they do anything less, since they will have
acted against the dictates of piety and justice? (46309^6)

54. Here one may cite a common thread, namely, a marked deemphasis on biology, run-
ning through Plato's treatment both of the family and of women's nature and possible role in
the ideal polis.
55. Cf. Bloom (1991, 385) for an emphasis on the conventional foundation of the family.
124 THE ANCIENT QUARREL

Following the tragic poets, Plato assumes that it is possible—and in crucial in-
stances desirable —for people who lack biological connections to one another
nevertheless to establish and sustain a whole series of kinship relations (463^6).56
Even as Plato builds in crucial ways on literary practice, he also diverges
from it in key respects. Above all, one may cite the metaphysical foundation
of his discussion, which ultimately governs all of the Republic's major innova-
tions. Most briefly put, the connections are as follows: The ground of every-
thing, once again, is Plato's conviction that Forms constitute natures (phuseis)
proper. Using his view of reality as a foundation, Plato divides human nature
into types, evaluating each based on its capacity to apprehend these natures.
Those individuals judged most favorably are in the best position, due to their
understanding, to inject order into the realm of Forms' participants. If they
are to be motivated and prepared for this undertaking, through their upbring-
ing and education it must become "second nature" to them to view the state
as though it were a single, living organism and to treat the "health," namely,
the order and unity, of that entity as their paramount concern. This can hap-
pen only if they are not distracted by private concerns, whether these be emo-
tional attachments or worries-about the accumulation and loss of wealth. As
Plato argues in Republic 5 prior to his discussion of kinship ties, both women
and men can serve as philosopher-rulers and, more generally, as guardians. 57
In his view, members of both sexes are best able to perform their social func-
tions if marriage and procreation are regulated so as to promote a social and
political unity that is far more than the mere sum of its parts. 58 Hence, having
started with the theory of Forms, one arrives eventually at those prescriptions
formulated at Republic 46id-e.
In addition, while one may extract generalizations about poets' views from
their remarks about functional terms and their referents, these generalizations
are not offered by poets themselves, who remain content to analyze the phenom-
ena on a case-by-case basis. Plato's metaphysical framework not only lends his

56. Plato and the poets differ, however, in their views about the composition of the two groups
and in their reasons for making assignments to them. For Plato, the crucial distinction is between
guardians and producers, since his proposal in Republic 5 does not include the latter.
57. For a recent defense of this claim about women's role as philosopher-rulers, see Levin
igg6b.
58. For Plato's emphasis on the close relation between the first and second "waves" discussed
in Republic 5, sec 45yc-d. The community (koinonia) of wives and children among the guard-
ians is identified as a crucial cause of unity in the polis (4643); as Vlastos (rgg5b, rj6) has argued,
the fact that Plato mentions women, but not men, in this context does not support the view that
men and women belong to one another in different senses.
 CONCEPTIONS OF APPROPRIATENESS 125

inquiry a systematic character absent from poets' reflections but also constitutes
the backdrop and justification for all of his proposals. Furthermore, in contrast
to tragedy, which concentrates on kinship relations, Plato's theory emphasizes a
close connection between the treatment of social roles — most notably ruling—
and matters of kinship; hence, his assessments of fitness occur with respect to
both sets of functional terms.
It is worth noting, finally, that there may be a link between Plato's treatments
of eponymy and the application of functional terms insofar as those properly
called "rulers" are viewed as such due to their awareness that a key range of terms
have primary and secondary referents. In Republic 5, Plato lays great weight on
the distinction between philosopher-rulers and those philodoxoi (4803) who are
merely enamored of appearances:

I distinguish accordingly those about whom you were just now speaking, the lovers
of spectacles, lovers of crafts, and practical people, from those who are the actual
focus of our argument and who alone merit the appellation "philosophers."
(47639^2; cf. 4803)

He then identifies the philosopher as

the one who believes that there is something beautiful in itself and can discern
both it and the things that participate in it, never mistaking its participants for it
or it for its participants. (47609^3)

To solidify this division between philosophers and all others, Plato articulates a
systematic distinction between faculties and their proper objects, emphasizing
that, unlike Forms themselves, participants in Forms (i.e., the objects of opin-
ion) are both F and not-F because they always partake (hexetai) qualifiedly of
the characteristics in question (4793-^. As previously discussed, Plato introduces
eponymy in the Phaedo explicitly to provide the semantics of the Form-participant
relation, making clear there that philosophers—as the only ones with knowledge
of Forms—will be the sole individuals who understand that terms referring to
Forms have a twofold application and hence how to make it properly. Thus, not
only does Republic 5 introduce a new dimension in Plato's discussion of appro-
priateness against a literary backdrop, but it may also take Plato's reaction to lit-
erary sources one step further by underscoring an essential relation between his
own two revised constructions of literary practice.
126 THE ANCIENT QUARREL

3. Conclusion
The foregoing inquiry stresses Plato's engagement with literary sources in three
central areas —etymology, eponymy, and the use of functional terms —as he
develops his philosophy of language, alongside the theory of Forms, in the middle
dialogues. As we have seen, this recourse to literary precedents issues, in both
the Phaedo and Republic, in revamped versions of notions prominent in litera-
ture. While Plato finds it most fruitful to rely on philosophical sources in the ar-
ticulation of his theory of Forms, I hope to have shown that when it comes to
crucial developments in his philosophy of language, it is above all to literary
sources that he turns. By bringing the latter debt to the fore, this study supple-
ments earlier accounts of the historical sources of Plato's theories. In particular,
by exhibiting his threefold concern with literature in the development of his
philosophy of language, it indicates that Plato's occupation with this anteced-
ent is far more extensive and thoroughgoing than previously recognized.
In the final chapter, I turn again to the Republic, concentrating this time on
Plato's most frequently discussed treatment of literary sources, namely, that pro-
vided in Books 2-3 and 10.1 maintain that—notwithstanding appearances to the
contrary generated by his sharp critique —Plato retains an important place for
the well-educated poet in his Republic. As I will argue, this account both fits the
text best and accords with Plato's own practice, in theorizing about aspects of
language, of acknowledging poets' contributions as warranted by their merit.
 5
THE QUARREL BETWEEN
PHILOSOPHY AND POETRY

A Reexamination of Poetry's Role in Plato's Republic

i. Introduction
Plato's opposition to poetry in his treatment of human character in Republic 2-3
and 10 has prompted extensive scholarly debate. Prominent topics of inquiry have
included the link between the dialogue's early and late reflections, discussions
of which often center on the concept of mimesis, and the question of what func-
tion Plato ultimately reserves for poetry.' Commentators have also treated the
connections between Plato's account of poetry in the Republic and views of it
expressed in other dialogues, particularly the Apology, Ion, Meno, Symposium
Phaedrus, and Laws.2 What has received far less attention, however, is the role
played by the Cratylus in Plato's critique of the literary tradition, and hence the
relation between his handling of poets there and in the Republic.
As indicated in chapter 2, accounts of the Cratylus' sources that emphasize
sophists and philosophers are problematic due to the paucity of extant evidence
and to the fact that such appeals are not clearly supported by the material that
has survived. As I argued in chapters 2-3, an investigation of literary practice,
combined with textual cues offered in the Cratylus itself, supports the conclu-
sion that this tradition is Plato's most direct and prominent target with regard to
etymologizing. For the purposes of the current chapter, which underscores ties
between his critiques of poetry in the Cratylus and Republic, the most salient

1. Regarding the concept of mimesis in the fifth century, see Else 1958; on the range of mean-
ings of "imitation" in Plato, one may consult McKeon 1936. For arguments that the accounts i
Books 2-3 and 10 are ultimately compatible, see, e.g., Tate 1928,1932,1938; Griswold 1981; Belfiore
1984; Reeve 1988, 220-31; Ferrari 1989; Asmis 1992. For a discussion that emphasizes a deep split
in Plato's thought about poetry in the Republic, see Annas 1982.
2. For discussion of these other dialogues, see Collingwood 1925; Tate 1928; Grube [1968]
1995; Schaper 1968; Ferrari 1989; Asmis 1992; Murray 1992.

127
128 THE ANCIENT QUARREL

consideration involving the former is its denial of the rank of techne to the liter-
ary practice of naming qua etymologizing due to its inability to meet the Gorgias'
criteria.
Far from limiting his critique of poetry's failure to qualify as a techne to its
treatment of naming, Plato argues, most concertedly in the Republic, that po-
etry itself does not meet the Gorgias' requirements. By so doing, he poses a di-
rect challenge to the familiar pre-Platonic assumption that its creators are the
practitioners of a techne.3 While interpreters have discussed at length the
Republic's objections to poets, especially in the arena of values, they have not
done full justice to the pivotal role of the techne framework in grounding its entire
critique. Comparative discussion of the Cmtylus and Republic with an empha-
sis on this framework reveals that the same conceptual threads govern Plato's
sustained critiques of poetry in what appear initially to be quite different arenas,
namely, philosophy of language and ethics. The inquiry thus offers further sup-
port to the view, defended in previous chapters, that Plato's exposition of the
theory of Forms and related doctrines pivotal to his version of philosophy has as
a close counterpart a systematic engagement with this rival "authority."
In addition to developing a multifaceted critique of poetry, Plato argues, above
all in Republic 5-7, that philosophy is the preeminent techne and that for this
reason the direction of education belongs properly to it. Plato situates his posi-
tive characterization of philosophy against the backdrop of his negative account
of rival occupations; in his view, the praise of philosophy and demonstration of
its rivals' inadequacies are essential and complementary aspects of a single project.4
The position to be defended here thus differs from that of Hall (1971, 204), ac-
cording to which the Republic's "relation of the philosopher rulers' 'craft' to the
world of forms and the idea of the good . . . distinguish [es] statesmanship from
the realm of technai." In my view, the cementing of this relation, above all to
the Good, is precisely what allows philosophy to meet in a superlative manner
those conditions on technai elaborated in the Gorgias.^
Following a discussion of philosophy's status, I consider what role remains
for poetry in Plato's ideal community. The account presented here differs both
from those that make philosophy and poetry the job of a single practitioner and

3. Regarding this assumption, see Murray 1996, 8; 1981, 98-99.

4. On the reciprocal connections between the critical and positive aspects of Plato's reflec-
tions, cf. Nussbaum 1986; Janaway 1995; Nightingale 1995. Based on the framework and discus-
sion of the Gorgias and on Plato's remarks elsewhere (e.g., the Protagoras and Republic), one can
develop a similar account of his challenge to the techne status of rhetoric and sophistry. The elabo-
ration of this challenge, however, is a topic for another occasion.
5. Hall (1971, 215-16) maintains that the Gorgias introduces "a transcendent ordering prin-
ciple" that "foreshadows the Republic's form of the good." While the Gorgias does look ahead to
the Republic's handling of goodness, I cannot agree with Hall's broader claim in the article that
the Gorgias is already moving away from the techne model, a process that Hall believes is com-
pleted in the Republic.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 129

from those that construe Plato as expelling poetry proper from his Republic.
While Plato rejects the view that poets are authorities in the sphere of pedagogy,
denying thereby that poetry could be a techne even under ideal conditions, he
admits the possibility that gifted practitioners, if themselves properly educated,
may benefit the state by generating creations that will be suitable for the young.
Although the pedagogical function of poetry is limited to children, poetic com-
positions will, in addition, play a civic role on a range of public occasions. Plato
thus "wins" for philosophy the quarrel (diaphora) between it and poetry by
arguing, against tradition, that philosophy should be the teacher of adults and
hence supplant poetry in this way as the educator of Greece.6 He establishes
philosophy's dominance, in addition, by contending that its practitioners should
be the ultimate arbiters of the content and form of acceptable poetic composi-
tions. The interpretation developed here foregrounds the complexity of Plato's
attitude by stressing that although much is at stake for him in the conduct and
outcome of the diaphora between philosophy and poetry, he is nevertheless
unwilling to bar poetry from the ideal polls if this exclusion would deprive it of
a potential benefactor.

2. Poetry and the Concept of Techne:

The Cratylus Briefly Revisited
The Cratylus focuses on one aspect of the content of poetry, namely, etymologi-
cal analyses of onomata that are supposed to disclose their referents' natures.
Plato's critique emphasizes that literary practitioners operate with belief, not
insight, and that their conception of "nature" (phusis) is fundamentally mistaken.
With respect to the latter, Plato rejects poets' assumption that individuals (e.g.,
Helen and Zeus)—whose names are subject most often to etymological analy-
sis—may be said to have natures in any direct or primary sense.
Because the understanding and subject-matter conditions are most promi-
nent in the Cratylus' challenge to the techne status of naming qua etymologiz-
ing, my discussion in chapter 3 of Plato's denial of this rank to literary practice
concentrated on them. 7 Yet insofar as etymologies may be used to promote
desired emotional reactions, the goodness condition would be quite pertinent
as well.8 Audiences' affective responses could indeed be shaped or enhanced by

6. For an argument that what Plato dubs a "long-standing quarrel" between philosophy and
poetry (Rep. 6oyb) was actually initiated by him, see Nightingale 1995, chap. 2; on this issue, cf.
Murray 1996, 231. For a different view, see Gould 1990.
7. For discussion of the understanding and subject-matter conditions with respect to the
Cratylus, sec chapter 3, section 2.
8. For a different view, which emphasizes the Cratylus' positive contribution to Plato's theory
of what art is, see Golden 1975,121-22. Although Plato concentrates in the Gorgias on showing that
rhetoric is a mere empeiria, he extends the results of that inquiry explicitly to tragedy, focusing thereby
13O THE ANCIENT QUARREL

appeals to names' descriptive content—a fact that Plato underscores in the

Cratylus. As we have seen, literary authors capitalize on this susceptibility in their
treatment of divine names such as "Apollo," which is traced to apollumi ("destroy").9
Productive of similar effects is writers' handling of many other onomata, among
them "Pentheus," which is tied to penthos ("grief), and "Helen," which is derived
from helein ("to destroy") in that famous passage from Aeschylus' Agamemnon (681-
92) discussed in chapter i.10 Putting the point in terms of the Republic's articula-
tion of the soul or mind (psuche), one may say that these and similar construc-
tions, which promote extreme feelings (e.g., powerful anger and grief) condemned
by the dialogue, foster the development of the spirited element at reason's expense.n
The presence of a concern with the goodness condition would, in addition,
help to make intelligible Plato's repeated emphasis in the Cratylus on the possi-
bility that words, and analyses thereof, can be manipulated to suit whatever
agenda the creator or user happens to have. The goodness condition is relevant,
moreover, insofar as those who follow names in their quest to understand reality
will be deceived, and thus come to harm, because they fail to identify and gain
proper access to those abstract entities, namely, Forms (eide), whose apprehen-
sion is the foundation of human flourishing.12 In fact, appeal to a concern with
the issue of benefit helps to explain why the Cratylus focuses so extensively on
etymology—a state of affairs that has prompted some to question whether the

on the goodness condition (5ozc-d; cf. Rep. 6oib). Regarding the claim's scope of application, it
is worth noting that in the Republic Homer is called the leader of the tragedians (5g8d); cf. 6o5c,
where he is identified as a maker of tragedy. On the reasons Plato and Aristotle viewed Homer in
this light, see Verdenius 1970, 212 n. 25.
9. Concerning "Apollo," see Aesch. Ag. 1080-82, 1085-86; see also Eur. Or. 119-21 and 954-56.
With respect to Greek tragedy, cf. Stanford (1983, 102), who states that "paronomasia could be pro-
foundly serious and pathetic in antiquity," noting that its emotional impact is most powerful when
based on a proper name. Regarding tragic poets' efforts to evoke and reinforce the emotions of eleos
and phobos, see Verdenius 1983, 50-51. For Plato's stress on the fear elicited by names, one may
consult his treatments of "Hades" (4033, 4O4b), "Persephone" (4040^), and "Apollo" (405d~4o6a;
cf. 4O4C5~7 and d8-e2). As he emphasizes in the Republic, names with negative descriptive content
are actually "designed to make everyone who hears them shudder" (387^ tr. Waterfield [1993]).
10. On Helen's name, cf. Eur. Andr. 105-6. Regarding "Pentheus," see Eur. Bacch. 508; cf. 1244.
11. On strong anger as a key feature of the tragic hero, see Gould 1990, chaps. 15,17-20. While
Plato's tripartite division of the soul into the rational, spirited, and appetitive aspects is not pre-
sented until Book 4, he emphasizes in Book 3 that poetry must not include strong expressions of
emotions such as fear and grief. For the claim that Plato assumes a "standpoint of perfection" in
his assessment of emotions in Books 2-3, see Nussbaum 1986, 157-58.
12. One will also be deceived if one does not realize that a key range of terms have primary and
secondary referents, namely, Forms and their participants, and that every relevant entity in the spatio-
temporal realm is an instance of the latter type; this point emerges above all in the Phaedo, Sympo-
sium, and Republic. For discussion of the Phaedo, see chapter 4. In the Symposium, one may con-
sult 2ioa-2i2a. Regarding the Republic, see, e.g., Plato's contrast between the lover of spectacles and
the philosopher (476-480) and his comments in connection with the Cave (5i5b-c) on the prevalent
error of assuming that entities in the realm of appearance are terms' exclusive, hence genuine, nominata.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 13!

discussion has a serious purpose.13 Even interpreters who grant Plato's earnest-
ness there have given voice to frustration, as when Elias (1984, 67) contends that
at times "the [dialogue's] reader is likely to bemoan the recurrent flogging of a
dead horse, above all in the etymologies, where a few examples quite adequately
illustrate the point, and a couple of hundred are hardly needed." While I agree
that Plato's intent in this portion of the Cratylus is serious, I also believe that the
length of his investigation is integral to its message. If Plato's intent were to show
only that those he targets lack understanding and associate natures in the pri-
mary sense with the wrong type of entity, he could easily have made these points
without producing such a lengthy catalogue of analyses. That the list is so exten-
sive, and spans so many different types of entities—all of which play a crucial
role in human life, albeit from varying points of view—suggests that Plato has a
broader aim in mind. 14 I propose that his goal, at least in part, is to emphasize
how harmful it is to one's well-being (i.e., the condition of one's psuche) to
approach the cosmic and human domains through the lens of etymology.

3. The Republic's Challenge to

Poetry's Techne Status
Various commentators, among them Collingwood (1958,18-19), have claimed
that Plato actually grants poetry the rank of techne.15 According to Hartland-
Swann (1951, n), Plato subscribes, not only in the Republic, but already in the

13. While I do not mean to exclude all humor or irony here, I do not believe that explana-
tions with this emphasis provide an adequate justification for the section's length and scope.
14. The foundation of Plato's inquiry is the investigation of divine and heroic names. Having
dealt extensively with these names, Plato moves on to onomata of other types of entities. Promi-
nent in 4o8d~4ioe are terms designating elements (e.g., fire), which are granted various types of
priority by Presocratics and are central to explanations of what they view as problematic. At issue
in 4113-4210 are so-called fine names (kala onomata) (41132, a8-bi). This class includes a range of
terms whose referents are of great importance to Plato in the areas of ethics, psychology, episte-
mology, and metaphysics. Key here is the fact that Plato includes terms denoting entities that he
takes to have the highest ontological status and to enjoy natural, logical, and explanatory priority
as well (agathon, dikaion, and kalon).
15. Having ascribed to Plato and Aristotle the assumption "that poetry . . . was a kind of craft,"
Collingwood proceeds to label that type of theory "a vulgar error" (1958, 18-19). 'n n's treatment
of the artist as a technites, Grey (1952) relies heavily on the Sophist, which, in my view, should
not be invoked here since that late dialogue relaxes the Gorgias' standards on what qualifies as a
techne (as evidenced, e.g., by its claim that sophistry has this rank), without, however, altering
the fundamental bases of Plato's critique. Moreover, while I concur with the emphasis of Schaper
(1968) on Plato's denial to poetry of the rank of techne, I cannot agree with her further claim (31-
32) that this challenge to its status is merely indirect. For a discussion that notes Plato's denial to
poetry of the rank of techne, concentrating thereby on its failure to meet what I call the "under-
standing condition," one may consult Woodruff 1982^ For the techne issue as central to Plato's
handling of poetry, see Janaway (1995), whose investigation concentrates on the Ion, Gorgias,
132 THE ANCIENT QUARREL

Gorgias (502) to the theory that poetry has this status. I maintain, in contrast,
that while the Cratylus deals with one portion of poetry's content via the techne
framework, the Republic rejects the techne status of poetry as such. Although
aspects of Plato's critique are familiar from other dialogues, the Republic's com-
prehensive challenge necessitates their adaptation and supplementation.
Plato's critical account of poetry in the Republic attends repeatedly to the
goodness, understanding, and subject-matter conditions on technai. The good-
ness requirement emerges as prominent already in Book i, where Plato stresses
that the activity of each techne has as its sole aim to do what is best (to beltiston)
for its objects (34731-3).16 Turning in subsequent books to poetry, Plato gives fre-
quent expression to the view that its creators aim to produce pleasure in the audi-
ence without a concern for what is beneficial. Tackling the problem on a specific
level, he maintains that passages in the writings of Homer and other poets that
promote a fear of death will be expunged, not because they are not pleasing (hedea)
to the audience, but because they do not benefit those who are to become guard-
ians ^Sybi-j). Lines that describe heroes as indulging too much in food and drink
will also be deleted: while they do indeed yield pleasure (hedone) in their hearers,
they fail to promote the development otsophmsune (3896-3903). Plato's concern,
most generally put, is that if familiar epic or lyric poetry is admitted into the city,
pleasure (hedone) and pain (lupe) will govern it rather than a concern for what is
best (Goya).17 Reflection on existing poetic constructions, whose effect on human
character he views as deeply problematic, leads Plato to conclude that if poets of
the familiar sort requested admission to the ideal state, they would be acknowl-
edged as "sacred and marvelous and pleasant (hedun)" but refused entrance for
the sake of the city's well-being (ophelias heneka) (jgSa-b; cf. 39yd). Any poet who
is to gain admission would have to show that his poetry "is not only pleasant (hedeia)
but also beneficial (ophelime)" to the community (6oyd). If poets' capacity to pro-
mote the good can be demonstrated, they will be admitted into the ideal polis.
Failing that, regardless of how much gratification they produce, they must be
excluded in order to prevent them from having a negative impact on its well-being.
In Republic 2-3, Plato indicts poets for harming others by treating as benefi-
cial that damaging state of affairs in which one has a reputation for morality but
is actually immoral. Here Plato traces poets' failure to meet the goodness crite-
rion to their shortcomings with respect to the understanding and subject-matter
(peri ti) conditions. Thus, he states (3636—3640) that, according to poets, licen-

Hippias Major, Republic, and selected later dialogues; while our arguments accord in their dis-
tinction between narrow and broad senses of techne (Janaway 1995, 47, 51), and in their conten-
tion that the Gorgias and Republic deny techne status in the former sense to poetry, they exhibit
significant differences in their structure, emphases, and details.
16. For the context, see ^6d$-34731. Other relevant passages are 342134-611 and ^bg-ez.
17. Concerning the dichotomy between the pleasant and the good insofar as it involves poets,
see also 493C-d.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 133

tiousness and injustice are only disgraceful in opinion (doxei) and by conven-
tion (nomoi), implying that poets promote this view because they lack under-
standing and the right conception of what is natural (phusei). Shortly there-
after, Plato attributes poets' promotion of a defective conception of flourishing
to their preference for appearance (to dokein) over reality (^6^o-c).
Republic 5, which presents a systematic distinction between knowledge and
belief, above all by differentiating their objects, builds on that division articu-
lated previously in the Gorgias (454). Books 6 and 10, which presuppose the
careful analysis of Book 5, comment most directly on how poets fare with re-
gard to their level of cognition. In Book 6, Plato stresses the prevalence of moral
deficiency by likening the populace to a powerful beast (493) and links together
poets and sophists with respect to their destructive impact (c-d). A negative ef-
fect on the public is inevitable as long as these^'strange teachers" (cy-8) operate
with and beget mere opinions regarding values.18 Plato supports his criticism of
practitioners' cognitive state by stressing individuals' inability to exhibit their
alleged wisdom in a logos of the procedures they employ and their relevance to
the end in question (493b-c).19 Regarding the subject-matter requirement, Plato's
repeated contrasts earlier in Book 6 between the activity of the few that is di-
rected toward Forms and the activity of the multitude that centers on partici-
pants (484b-e, 4853^, 49oa-b) provide the foundation for the beast analogy.
That the analogy is followed immediately by the repetition of this contrast (493C-
4943)—which itself begins with Plato's request that one bear in mind what has
just transpired (49362)—suggests that specialists' failure to transcend appearance
promotes the audience's obsession with its manifold permutations.
In Book 10, Plato concentrates on poets, repeatedly emphasizing their lack
of insight (598d-6o2b). While some say that poets know all technai (jgSd-e),
Plato insists that they know none. One cannot reasonably hold them to account
the way one would doctors and generals when they discuss medicine and
generalship, respectively. One must, however, hold poets responsible for the
negative impact of their constructions where the issue is virtue and vice, since it
is on poets' handling of such topics that their claim to excellence rests. Although
poets allege to have mastery of these subjects, in fact they lack knowledge of them
altogether (6ooe-6oib). Generating only products "three removes from reality"
(5976), poets deceive —and thereby corrupt—the audience by manipulating
words (onomata) and phrases (rhematd) to create false impressions of expertise.20
Further discussion of Book 10, which deepens Plato's critique of poetry with
respect to all three techne conditions, will be provided in section 6.

18. Regarding Plato's critical stance toward the production of beliefs by nonphilosophers, cf.
4963.
19. For an emphasis on the importance of being able to provide a rational account (logon
didonai) that exhibits one's understanding, see also 5330 and 534b.
20. On poets' deceptive use of onomata and rhemaia, see 6oia—b.
134 THE ANCIENT QUARREL

The Republic's negative assessments of poets with reference to the understand-

ing and subject-matter requirements are familiar from the Ion and Cratylus and
supplement in crucial ways those offered in the latter.21 Both the Ion and the
Cratylus stress poets' lack of understanding. The Ion not only emphasizes the
fact that poets operate without knowledge but actually treats the absence of nous
as a necessary condition for creative practice (53/fb). 22 In addition, and in con-
trast to the Republic, the Ion's investigation stresses a contrast between knowl-
edge and divine dispensation rather than one between knowledge and belief—
which may be true or false, and true for quite different reasons.23 The Cratylus,
in turn, as we saw, condemns poets for their lack of understanding as evinced by
their attempts at naming, which are judged to be products solely of opinion. Al-
though one can certainly apply more generally Plato's basic criticism there that
poets operate with mere belief rather than knowledge, one must consult the
Republic for that more general treatment, along with a sustained exploration of
the relation between the two cognitive states.
The Republic's fundamental objection to poets with respect to the subject-
matter condition is that they are occupied not with ethical realities themselves,
for example, the nature of justice or courage, but with semblances (or semblances
of images) thereof.24 This objection is similar in type to that developed in the
Cratylus but is significantly broader in scope. The common ground lies in the
fact that in both cases Plato criticizes poets for having the wrong conception of
phusis, one that makes particular entities fundamental and often exemplary.
Regarding the issue of scope, while the Cratylus focuses on mortal and divine
individuals—which is to be expected since literary etymologies are largely of
proper names—the Republic's critique of poets is formulated with respect to a
much broader range of participants in the relevant Forms: here the entities in
question include not only individuals possessed of agency but also actions (e.g.,
those alleged to manifest bravery or self-mastery) and institutions (e.g., the prac-
tice and social role of the hero as described by Homer). 25 Plato's point here is

21. See Janaway (1995, 16; 1992) for a different view of the Ion, according to which Plato does
not deny, but rather assumes, the existence of a poetic techne.
22. The Ion does not specify what operating with understanding would amount to. What Plato
does say is that knowledge is of a determinate and clearly circumscribed subject matter, whose
ontological status is not specified. He also identifies one consequence of having knowledge,
namely, that one can exert oneself to extend the sphere of one's expertise.
23. With respect to the different grounds of doxa, see Rep. 429^430^ where Plato distinguishes
true belief produced by education from that generated by any other source.
24. For a challenge to the familiar view that paintings and poems are metaphysically distinct
from those objects that creators represent through their activity, see Belfiore 1981,120 (with n. 19).
In her view (120), a painting of a bed "is qualitatively rather than metaphysically different from
the bed we use, for it is made by an incompetent individual who is capable of understanding and
imitating only apparent beauty."
25. While the Cratylus itself, particularly the conclusion, emphasizes the broad scope of Plato's
concern, it is in the Republic that the account is generalized explicitly with reference to poets.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 135

that, notwithstanding poets' captivating descriptions of entities of these types,

none of them may be taken as fundamental or paradigmatic.
Finally, both the Cratylus and the Republic evince a concern with the good-
ness condition. In the Cratylus, this worry—which is not totally explicit—cen-
ters on the potentially destructive impact of names with negative descriptive
content and, more generally, on how harmful it is to the welfare of one's soul to
use etymologies as vehicles for apprehending "reality." The Republic's concern
with poetry's impact is, in contrast, quite direct. As we saw in chapter 2, Plato
forges here a strong link with ground covered in the Cratylus by emphasizing
the destructive potential of names with negative semantic constitution and by
noting that the incorporation of onomata with positive descriptive content in
revised poetic compositions could be a fruitful pedagogical tool (Rep. jSyb-c).
In addition to the fact that its concern with the goodness criterion is fully expli-
cit, the Republic differs from the Cratylus insofar as Plato's efforts in the former
are directed toward poets' constructions in their entirety and hence yield a con-
clusion that applies to their practice as a whole. The conclusion to draw from
the Republic's thoroughgoing critique of poetry is not that virtue itself is not the
province of a genuine techne. It is indeed the domain of a techne, but that techne
is philosophy rather than poetry.

4. Plato's Defense of Philosophy

as the Techne Par Excellence
Plato's most concerted attempt to depict philosophy as being at the top of the hi-
erarchy oftechnai comes in Republic 5-7.26 There he makes clear that it not only
satisfies all the conditions set forth in the Gorgias but meets them to the fullest
possible extent. Based on his exploration of the subject matter of philosophy, the
cognitive state of its practitioners, and their superlative ability to meet the good-
ness requirement, Plato concludes that philosophy is the preeminent techne.
Much controversy surrounds the role of techne in Plato's theorizing about
philosophy. A key point of contention is how one should interpret the craft anal-
ogy presented in early dialogues such as the Laches and Charmides. Should one
construe the argument to be that arete ("virtue," or "excellence") itself is a
techne?27 Might the connection between techne and virtue be intended more
loosely, such that while the comparison gives some positive content to a moral
theory, difficulties and disanalogies are evident even at this early juncture? 28 Or,

26. On philosophy as having this position, cf. Kato 1986, 47, 72, 77.
27. See Irwin (1977!}, 7), according to whom Socrates "argues that a virtue is simply craft-
knowledge." Warren (1989) prefaces his case for the view that the Republic depicts ruling as a
techne by claiming (101) that, strictly speaking, "the craft argument is not an analogy at all but is
an expression of Plato's conviction that knowledge is virtue."
28. Regarding this position, see, e.g., Vlastos 19953; Gould 1955; Schofield 1979; O'Brien 1967.
136 THE ANCIENT QUARREL

in contrast, is the craft analogy used to criticize, refute, and exhort, with the
knowledge involved in virtue to be construed as nontechnical? 29 Since Plato's
early use of the craft analogy has been thoroughly discussed by others, and the
current study concentrates on subsequent dialogues, this issue will not occupy
us here. For present purposes, I simply note my adherence to the view that while
this analogy sheds some light on how virtue should be understood, there are
difficulties with it, which stem from unresolved questions about the relation of
technai to the good and their restriction to particular domains in a way that is
not entirely appropriate to virtue.30 Such difficulties are not, I believe, irremedi
able. In contrast to Gould (1955), I suggest that Plato's later reconstruction of
techne avoids just these problems.31
Most relevant for current purposes is the fact that those who grant positive
import to the early dialogues' techne analogy disagree about what happens sub-
sequently. Some commentators argue that Plato, having seen clearly and taken
to heart the inadequacies of the craft analogy, is prompted to reject it.32 Others

29. For this view of the analogy, see Roochnik 1996,125 (cf. 1986, 303, 307; 1990,187-88,193);
concerning the nontechnical status of knowledge, see Roochnik 1996 (cf. 1990, 93, 190-91). The
claim that the knowledge in question is nontechnical is also endorsed by Gonzalez (19983, chap.
2), who cites Roochnik's interpretation of the analogy's function approvingly (294 n. 47). Central
to the question of how the argument should be construed is the matter of the relation between
techne and instrumentality. Irwin (igyyb) argues that in the early dialogues virtue is construed as
merely instrumental to happiness, and therefore as lacking in intrinsic value; see also his more
recent affirmation that "an instrumental view of virtue and happiness . . . fits the dialogues best"
(1995, 76). For challenges to the instrumentalist thesis, see Vlastos 19953, 131; Taylor 1979, 599;
Zeyl 1982; Nussbaum 1986, 97-98; Lesses 1986; Penner 1992,149-50 n. 14; Parry 1996,101-4. Ther
is disagreement, in addition, regarding the extent to which Socrates breaks new ground in turn-
ing to the craft analogy. Gould (1955, 31 n. 3) opposes his own emphasis to that of Robinson (1953,
206) by claiming that Socrates "creates the analogy" (cf. 33); cf. Irwin (igyyb, 289 n. 20), who
contends that "the connection between virtue and craft is not at all intuitive to Greeks." On the
argument of Kube (1969), there is continuity between the techne analogy used by the Platonic
Socrates and earlier Greek ideas; see O'Brien (1967, chap. 2) for the contention that an existing
view of politics and virtue as a techne is a major source of Socrates' and Plato's "ethical intellec-
tualism." On pre-Platonic usage of the term techne, see also Schaerer 1930; Roochnik 1996.
30. For discussion of issues involving the goodness and domain of technai, see Gould 1955,
chap. 2. Roochnik (1996) emphasizes difficulties with the techne model that center on the good-
ness and subject-matter conditions, but he differs from Gould in believing that the dialogues'
model with respect to virtue is always nontechnical.
31. Regarding the subject-matter criterion, see Hogan (1977), who argues, contra Gould, that
the Charmides does not disclose problems with the analogy per se but instead illustrates the dia-
lectical impasse that prevails as long as the nature of the object in question remains unspecified.
For critical discussion of Gould on the issue of what has been called (e.g., by Roochnik 1996, 30)
"value-neutrality," see Tiles 1984, 51-54. Like Gould, Roochnik (1996; 1990, 190) maintains that
technai are inherently value-neutral.
32. See, e.g., Irwin ig77b, 203: Plato "denies that moral knowledge is a craft prescribing in-
strumental means to a previously desired goal" (cf. 248); Irwin notes subsequently that in the middle
dialogues "moral knowledge is episteme, but not techne" (289 n. 19). For further discussion of the
Republic and a renewed defense of the view that Plato rejects the craft analogy, see Irwin 1995.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 137

contend that Plato retains a positive role for the concept of techne while intro-
ducing revisions that address those difficulties associated with the early dialogues'
construction of it.33 In what follows, I defend my own position, which may be
described as a version of the latter stance.34

On this issue, cf. Collingwood (1958, 18), who interprets Rep. 5^1-3363 as demonstrating that
justice is not a techne, and the claim of Cross and Woozley (1964, 13) that Socrates' argument
against Polemarchus at Rep. 333-334 "shows why it is a mistake to regard justice as being a skill
at all." Although on balance Hall (1971) seems to belong in the group of those who claim that
Plato gives up the craft analogy (216; cf. 205), his comments sometimes suggest instead a distinc-
tion within the realm of technai, as when he claims that in the Republic "statecraft is not one
among many technai but in some significant respects unique" (214). Woodruff (1990) attributes special
importance to the Meno's recollection theory, and distinction between knowledge and true belief,
arguing that "a consequence of these developments is that the fecrme-model for knowledge is aban-
doned, for here Socrates considers a sort of knowledge that is always present in the knowcr, and so
never taught" (8r). In my view, Plato's introduction of the recollection theory need not, indeed does
not, jeopardize an activity's capacity to meet the learnability/teachability condition on technai;
instead, one simply requires a model of the content, goals, and means of teaching that corresponds
to this new conception of learning. Plato, I believe, offers precisely this in the Republic.
33. For emphasis on the positive role of the concept of techne in Plato's mature moral theory
and treatment of the philosopher-ruler, see Sprague 1976; Parry 1983, 1996; O'Brien 1967; Kato
1986; Kube 1969; cf. Nussbaum 1986 (esp. 237, 290-91) and Wild 1946. One may consult Warren
(1989) for the view, based above all on the arguments of Republic i, that ruling is the "mastercraft."
Thayer (1969) emphasizes the positive role of techne in Plato's philosophy, but his contention
that "the polls" is the preeminent techne in the Republic (259, 260) is insufficiently precise.
Bambrough (1971) sees continuity between Socrates and Plato in the view that the focus of in-
quiry, virtue or statesmanship, is itself a techne; according to Bambrough, by depicting statesman-
ship as "prescriptive" rather than "instrumental," Plato tries unsuccessfully to address the diffi-
culty that qua techne practitioner the statesman would formulate means not ends (201-2). For an
argument that Plato, having seen the craft analogy's fruitful application to Socratic ethics, uses it
in the Timaeus as a model for natural science, see Graham 1991.
It is not entirely clear which of the two camps Reeve (1988) fits into. While he stresses that
Plato rejects the craft analogy (see, e.g., xii, 8, 19) —leading Roochnik (1996, 5 n. 14) to identify
him as a "discontinuist"—there are indications that Reeve does attribute a positive role to techne
in Plato's mature theorizing. Thus, for example, having claimed (23) that Plato "does not aban-
don [Socrates], or his theories; he transforms them," he contends that philosopher-rulers prac-
tice a craft, indeed, qualify as "master craftsmen" (84-89). That Reeve views Plato as distinguish-
ing levels of crafts receives support from Reeve's distinction (88) between the "ordinary craftsman"
and an "extraordinary" one, which is followed by the assertion (89) that Plato's description (596^2-
di) of the extraordinary craftsman "fits the philosopher-king like a glove." Since philosophers —
who are unique in having genuine knowledge "about virtue, and what kind of life is best" (Reeve
1988, 90)—are said to practice a craft, it is hard to avoid the conclusion that Reeve is at least a
qualified advocate of continuity.
With respect to techne, cf. Kahn (1996), who stresses the importance of this concept to Plato's
theorizing about morality and politics in dialogues up through the Republic in the context of
defending "the essential unity of Plato's thought" (xvii). Contesting the dominant view, Kahn
argues that Plato's early and middle dialogues do not provide evidence of "a distinct Socratic
period" (39), manifesting instead a single Platonic vision that is gradually articulated.
34. The account of philosophy to be sketched here diverges sharply not only from the stance
of those who claim that Plato gives up the techne framework but also from the position of Roochnik
138 THE ANCIENT QUARREL

In the early and middle dialogues, Plato distinguishes repeatedly between

seeking merely to produce pleasure in others via flattery and truly promoting
the good of those toward whom one's activity is directed. Plato's argument in
the Republic bases philosophers' distinctive capacity to promote their own and
others' welfare on their apprehension of the highest element of reality, namely,
the Form of the Good.35 In Plato's view, nothing imperfect (ateles), that is, noth-
ing that is not both unchanging and unchangeable, provides an entirely firm
reference point for either thought or action (5040). Prior to his discussion of the
Good, Plato insists that it is best for philosophers to rule in the community
because only they truly understand the elements of reality and their intercon-
nections. Philosophers assimilate themselves to the harmony that they have
glimpsed among these entities (5oob8-di), and this apprehension and resulting
accord within enables them to foster harmony without, once again taking as the
foundation of their activity Forms, particularly values, and their interrelations
(5ood, 5oib-e). Having established firmly this causal relation between the appre-
hension of perfect objects and the securing of benefit, Plato introduces the
capstone of his theory, namely, the Form of the Good—the ultimate object of
aspiration—which lies at the top of the hierarchy of Forms (see jc^d-^ogd, 5iyb-
c, 5i8c-d).36 While all human beings want to promote what they view as good,
particularly for themselves (50^-6; cf. 5213), only the apprehension of this Form
allows one to come as close to perfection in the structuring of self and commu-
nity as is possible for anything tinged with mortality (see especially 5053-506^
5ogb, 5143-5213). Philosophers' apprehension of the Good thus allows them to
maximize the presence in the spatiotemporal realm of what is genuinely bene-

(1996), who, as previously noted, maintains that the dialogues never seriously propound the craft
analogy. Roochnik focuses on the early dialogues but maintains that "if in [Republic] i and his
early works Plato was pointing to a conception of nontechnical moral knowledge, it becomes
increasingly likely that in his middle works he also eschewed a technical moral theory" (146;
cf. 135). This view of the middle dialogues need not follow, and indeed might be less persuasive —
a priori at any rate—to one who embraced a doctrinal approach to Plato's writings that attended
to the development of his theories over time. Roochnik's treatment of the early dialogues stresses
the interdependence of (dialogue) form and content, and his remarks on the figure of the "ship-
of-state" in Republic 6 (149-50) make clear that, were he to offer an account of Republic 2-10, it
would underscore this relationship. A key assumption on which Roochnik operates is that the
form of Plato's discussions tells against their content's being doctrinal. If one does not accept this
assumed connection, however, then Roochnik's "nontechnical" interpretation is deprived of a
key source of its support. For critical discussion of Roochnik's approach from a different perspec-
tive, see Weiss 1998.
35. Cf. Sprague (1976, 99), according to whom a vision of the Good grounds the Republic's
distinction between "legitimate and illegitimate second-order arts."
36. This claim is not intended to apply to all stages of Plato's reflection on Forms. The iden-
tity of those Forms or kinds lying at the top of Plato's classification does in fact undergo some
variation. Most notably, in the Sophist he gives primacy to the five megista gene. It is from the
perspective of a concern with values that Plato situates the Good above all other eide.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 139

ficial. 37 Practitioners of other technai, in turn, operate under the supervision of

the philosopher insofar as they require the latter's expertise in order to deter-
mine whether the goodness requirement is being met to the highest degree pos-
sible within their own domains of expertise.38 Although there is a special rela-
tionship between the Good and other ethical Forms, more generally Plato views
order and harmony as marks of goodness (for his assignment to the Good of far
wider than ethical significance, see 5086-5093).39 The concept of goodness is
thus pertinent to such disciplines as astronomy, which studies the ordered move-
ment of celestial bodies. There is a close connection, moreover, between the
cosmic and human planes insofar as the proper study of disciplines like astronomy
helps to create order and harmony within the human soul.
In the Republic, Plato emphasizes not only that technai are markedly supe-
rior to empeiriai but also that technai themselves are situated hierarchically with
reference to one another (see Plato's pointed distinction at 533b between phi-
losophy and "all the other technai" and his contrasting of philosophy to "the other
technai" at495d). As the Divided Line makes clear, the second tier of technai is
typified by mathematical activity, which involves the direction of one's atten-
tion toward eide, with, however, those Forms' being explored indirectly (i.e., via
particulars like diagrams) rather than firsthand. Plato's praise of it in Books 6
and 7 is qualified: insofar as mathematical practice is concerned ultimately with
Forms, it is to be lauded, while due to its reliance on particulars and unques-
tioning use of hypotheses, it is rightly subject to criticism.40 Above it is that activity

37. Much controversy surrounds the question of whether it is in philosopher-rulers' interest

to divide their time between philosophizing and ruling rather than devoting themselves exclu-
sively to the former activity. On this issue see, e.g., Kraut 1973,1992,1993; White 1979, 1986; Annas
1981, 259-71; Cooper 1977; Irwin i977b, 236-37, 242-43 (with 337-38 n. 6r), 267-68, and 1995,
chap. 18; Reeve 1988, 200-203. 'n mY v i gw > ruling does not detract from the well-being of these
individuals. As Kraut (1992, 328—29) observes, "one's highest good is not always served by purely
contemplating the Forms; rather, one's highest good is to establish and maintain a certain imita-
tive relationship with the Forms, a relationship that is strained or ruptured when one fails to do
one's fair share in a just community."
38. On the philosopher's role as overseer, cf. Kato 1986, 23, 47, 76. In the view of Sprague
(1976, 92), the "first-order" arts are introduced as "steppingstones to the good and the good then
descends to make them truly useful." According to Kahn (1996, 209), "the Good itself. . . must
be the object . . . for the art of the philosopher-kings. And such knowledge will be useful pre-
cisely because, in the hands of the rulers, it will guide the right use of the workings and products
of all the other arts."
39. On these features as signs of goodness, see also Kraut 1992, 323.
40. There are various ways of construing the difference between stages 3 and 4 of the Divided
Line. On the interpretation adopted here, the difference is not one of objects but rather of meth-
odology, insofar as the entities studied by mathematics are also at issue on the highest level. Another
stance would involve viewing the objects of stage 3 as mathematical entities, intermediate be-
tween particulars and Forms. Ross (1953, 6;) presents a third option, according to which "the
objects of Sidvova are not the 'intermediates' but are simply the mathematical Ideas, and those
of voi>5 the other Ideas" (see also Robinson 1953, 195). Although this is an interesting suggestion,
in my view Plato's handling of the Line and his comments elsewhere, e.g., in the Recollection
140 THE ANCIENT QUARREL

which Plato views as constitutive of philosophy. Here one proceeds solely via
Forms, and all movement from particulars to Forms is viewed as merely prepa-
ratory (for a description of this type of activity, see 5iib-c; cf. 533b-d).
According to the Line, while the inquiries of mathematicians and philoso-
phers both involve Forms, only the latter have immediate access to those enti-
ties combined with a systematic understanding of how they interconnect. This
comprehensive understanding is based ultimately on that knowledge of the Good
which sets the philosopher apart from everyone else. Hence, philosophy differs
most fundamentally from other technai with respect to its subject matter because
it alone has the Form of the Good as part of its domain. Since it is the apprehen-
sion of this Form that allows one to have the most direct and complete access
possible to other eide, along with full knowledge of their interrelations, its being
an element of philosophy's domain ultimately explains why philosophy has
Forms as its subject matter in the strictest possible sense.41
While Republic 5 elaborates the distinction between knowledge and belief
made by Plato in the Gorgias, it does not specify how technai such as geometry
are to be accommodated in terms of this dichotomy. This omission stems pre-
sumably from the fact that Plato is beginning his detailed exploration of how
philosophers differ, cognitively speaking, from everyone else.42 In his presenta-

Argument of the Phaedo (726-773), support the contention that the philosopher's activity involves
Forms for both mathematical and value concepts. While Robinson grants that "Plato probably
held that the objects now studied by mathematical method could be successfully studied dialec-
tically" (1953, 200), in my opinion a stronger claim about the actual domain of dialectical in-
quiry is justified.
41. Although, as commentators have observed, the Form of the Good is not mentioned ex-
pressly in the Line, it need not follow that it is irrelevant thereto. According to Fogelin (1971,
378), the Good has no place whatsoever there. I would stress, in contrast, that even if the Good
has special features that distinguish it from other Forms, it still belongs to the domain of eide and
therefore shares central characteristics with other members of that group (e.g., being abstract,
timeless, and such that entities in the spatiotemporal realm can participate in it). In addition, as
Robinson (1953, 159) observes, the Line assigns a unique position in human knowledge to the
anhupotheton (510^7, 5iib6), and there is evidence to support the conclusion that this entity is
none other than the Good (for justification of this inference, see Robinson 1953, 159-60).
42. At first glance, that portion of Republic 5 in which Plato makes his remarks about women's
equality does not appear to be tied closely to what follows. In the view of Grube (1974, rn), the
first two-thirds of the book "deal with subjects which have been omitted or but lightly touched
upon in founding the city, while the last third begins the description of the Platonic philosopher,
his wisdom and his methods, which continues through the next two books" (cf. Cornford 1941,
144, xii; Halliwell 19933, 3-5). If, however, one understands the discussion of women in terms of
its role in Plato's defense of philosophy as the techne par excellence, the structure of Republic 5
ceases to be problematic. As previously emphasized, the understanding requirement on technai
presupposes the existence—and delineation—of a small class of expert practitioners. In Plato's
assessments of activities' claims to the rank of techne, discussions of these aspects of the under-
standing condition go hand in hand and are mutually reinforcing. For example, in the Cratylus,
which denies techne status to naming qua etymologizing, Plato's handling of the understanding
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 141

tion of the Line ^ogd-jne), Plato claims that dianoia ("reasoning") is interme-
diate between opinion and understanding (noesis, a term used interchangeably
with episteme; for the latter, see 5336-5343). Like noesis, dianoia is directed toward
Forms, but its access to those entities is mediated rather than direct. When Plato
draws a general contrast between the visible and intelligible realms, which are
symbolized by the halves of the Line, he refers to the bottom two segments col-
lectively as doxa and to the top two as noesis (5336-5343). When he speaks with
the highest level of precision, however, a distinction within each half—most
notably the upper one — comes to the fore. The cognitive framework represented
by the Line makes clear that practitioners ottechnai that occupy the tier below
philosophy—"geometry, arithmetic, and so forth" (5100) —achieve something
short of full understanding. With regard to the understanding requirement, the
philosopher thus surpasses not only those who practice empeiriai but also all other
techne practitioners. 43
In my view, Plato resolves the difficulty with the techne analogy that involves
virtue's lack of a clearly delimited subject matter (along the lines, say, of medi-
cine or astronomy) by clarifying and refining the connection between virtue and
techne. While the Laches and Charmides are concerned ultimately with the
"What is F-ness?" question, the investigations of virtue themselves focus, often
interchangeably, on qualities, people, and activities. The question of whether
virtue is a techne in fact centers most directly, not on qualities, but rather on
praxis. Insofar as the latter is the focus of inquiry, the early dialogues find no
proposed candidate for its subject matter to be adequate.44 A full articulation of
the relation between virtue and techne would presuppose in the treatment of
this issue a clear division of qualities from activities of the sort that is not offered
at this early juncture. In the Republic, Plato cements the distinction between
qualities and an activity or way of life by delineating the latter as a techne, namely,
philosophy, and the former, construed as Forms, as the core objects around which

condition comprises his argument both that the activity lacks practitioners of the requisite cali-
ber and that those who have undertaken naming to date have operated based solely on doxa.
Republic 5 also treats these two complementary aspects of the understanding condition: the last
section, which presents a systematic distinction between knowledge and belief, addresses the is-
sue of what cognitive state distinguishes the philosopher, while the earlier portion, which con-
tains Plato's discussion of women, helps to specify the range of human beings who may count as
expert practitioners of philosophy under ideal circumstances. It is thus far from coincidental that
Plato presents his insights about women, which pertain directly to the issue of who may possess
this highest form of expertise, as a prelude to detailed remarks on the cognitive state in which
that expertise is manifested. On this interpretation, Republic 5 is a unity in a far stronger sense
than is captured by interpretations that treat the book as merely transitional. If one calls to mind
here Aristotle's discussion of substantial unity in Metaphysics Z.iy, one might say that Book 5 is
like the syllable rather than the heap (104^11-33).
43. On Book lo's presentation of artifact construction, see section 5.
/|/1 For discussion of the Charmides, see Hogan 1977.
142 THE ANCIENT QUARREL

that life is built. What is more, Plato's construction of goodness makes it pos-
sible for philosophers to maintain virtue as their special province while also appre-
hending the goals and domains of other technai, particularly insofar as the quali-
ties of order and harmony pertain thereto.45 Plato's conception thus shows clearly
how philosophers' understanding can be focused on the domain of virtue yet
consist, at the same time, of a vision that is all-encompassing.
Plato's revised construction of techne also addresses successfully the worry,
which is especially prominent in Republic i and in the Hippias Minor, that
expertise can just as easily be used to promote bad as good. One might think
that the provision of benefit is ensured by the fact that the Gorgias' framework
gives a crucial role to the goodness requirement. Actually, however, while the
goodness criterion captures Plato's insistence on barring the provision of harm —
whether due to malice or to culpable ignorance—from the realm of technai, it
does not itself explain how this exclusion is possible. What ultimately guaran-
tees techne practitioners' promotion of goodness in the spatiotemporal realm is
the fact that philosophers have direct cognitive access to the paramount object
of understanding, namely, the Good itself. This apprehension, in turn, prompts
them to act directly on the spatiotemporal domain so as to maximize the pres-
ence of goodness within it, and to advise practitioners of other technai on how
the goodness condition can be met most completely and effectively within their
own areas of specialization.46 Although it might seem questionable or problem-
atic that philosophers' apprehension of goodness could benefit, say, the astrono-
mer, Plato's strong connection of goodness to order and harmony makes this
link seem natural indeed.
The foregoing discussion of the Republic's treatment of poetry and philoso-
phy takes as its lens the concept of techne. In particular, it shows how in that
dialogue Plato both denies techne status in the narrow sense to the entire prac-
tice of poetry and argues that philosophy is itself the techne par excellence. The
depth of Plato's concern with the goodness criterion is manifest in the Republic's
sustained attempt to demonstrate that philosophy, along with those institutions
established and monitored by its practitioners, is of greatest benefit to human
beings. Since Plato's praise of philosophy and denunciations of its rivals tran-
spire alongside one another, one might anticipate — rightly—that his challenges
to the latter would emphasize their shortcomings with respect to this condition.
The account presented here, and expanded in section 6, suggests that Plato's
treatment of poetry follows that of philosophy in another respect as well; namely,
just as his defense of philosophy makes paramount its unique standing with
respect to the criteria involving subject matter and understanding, so too, con-

45. The connection between goodness and order is adumbrated in the Gorgias, where Plato
insists that a soul is in good condition to the extent that it manifests structure (taxis) and order
(kosmos) (Grg. 50^-507^ esp. 504!} and 5o6d-e).
46. Regarding the former point, cf. Parry 1996, 148-50.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 143

versely, at bottom Plato's critique of poetry is for its failure to meet these two
requirements.
Not only does this inquiry help answer the question of how Plato views the
relation between poetry and philosophy under current conditions, it also places
at least some general constraints on any future connection that might obtain
between them. Above all, we know that a valid association is possible only if
poetry's benefit to its hearers, and hence its contribution to human flourishing,
can be demonstrated. While one might think that the specifics of poetry's fate
will simply fall out predictably, and with a wholly negative outcome, from what
precedes, in fact the discussion is not yet complete. In particular, it is worth
addressing directly and in some detail the question of what future relation Plato
envisions between poetry and philosophy. The result will be a more positive view
of poetry's potential role in the ideal polls than the preceding emphasis on his
critique of it might lead one to suspect.

5. A Techne of Poetry? The Status

of Revised Literary Practice
As noted, Plato allows that there is a place for poetry in his Republic to the extent
that it can be shown to promote the optimal functioning of individuals and hence
the community. Defenses of poetry along these lines—whether offered in verse
by poets themselves or in prose by others—will be looked upon with favor because
"we gain if it turns out to be not only pleasurable but also beneficial" (60761-2).
The point of contention is whether and how poetry can be helpful in the struggle
to be good (6o8b), including the question of whether it can do so by meeting
Plato's central requirements on technai. I suggest that poetry can indeed be of
use in this struggle even if it is not, and cannot be, itself a tec/me.47
The most promising approach for one who wanted to defend the claim that
a revamped practice of poetry achieved the rank of techne would be to argue
that it was analogous in the relevant ways to carpentry as described in Repub-
lic 10. In my view, however, this line of reasoning does not yield the conclusion
that poetry is a techne in the strict sense that is delineated in the Gorgias.
Republic 7 includes atthe bottom of its classification of technai those "directed
toward generation and manufacture," 533b.48 Although Plato does not spell out
here what his criteria are for placing artifact production (e.g., bed-making) on
the lowest tier of this hierarchy, one may infer the assignment's foundation from
what he says elsewhere.

47. Cf. Woodruff (1982!}, 150 n. 12), who maintains that although Plato "entertains the idea
that a good and skillful rhetoric could be developed (in the Phaedrus), he holds out no such hopes
for poetry."
48. I adopt here Waterfield's translation (1993).
 THE
144 ANCIENT QUARREL

Bed-makers lack, not only knowledge, but also dianoia (i.e., the cognitive state
of mathematicians according to the Divided Line). As Republic 10 makes clear,
they operate instead with true belief (5g6b with 6oic-6o2a). Practitioners of
carpentry thus fail to satisfy the Gorgias "understanding condition," even on that
looser construction according to which it may be met by dianoia—& state that
falls short of insight, yet is distinctly superior to belief. The Gorgias places much
weight on the requirement that techne practitioners have understanding rather
than mere belief, and under ordinary conditions, an activity's failure to meet it
would suffice to exclude it from the realm oftechnai narrowly construed. In the
Republic, however, Plato tells us that bed-making and other relevant types of
artifact production do make the grade.
Practices like bed-making qualify as technai above all because they have eide
as their objects, fulfilling thereby the subject-matter condition. Yet the fact that
true belief is sufficient here for activity involving ideal should alert us to the strong
possibility that we are dealing with entities that are not strictly on a par with those
central cases of Forms delineated in the Phaedo (for Plato's focus there on enti-
ties in the realm of mathematics and of values, see, e.g., yjc-d): any access to
those Forms, as we have seen, requires at a bare minimum dianoia—and the
apprehension that it makes possible is merely indirect.
Plato's only explicit remarks in the middle dialogues about some Forms'
having an elevated position relative to others occur in Republic 6 and 7, where
he introduces the claim that the Good lies at the pinnacle of his classification.49
While Plato does not address directly the issue of whether there are distinctions
of level among the remaining Forms, there are indications that the eide involved
in artifact production reside on a lower tier than does the core group of Forms
familiar from the Phaedo.
In the Peri Ideon's "Arguments from the Sciences," Aristotle contends, osten-
sibly, that Plato's account commits him to the existence of more Forms than
Plato himself actually wants; most specifically, Aristotle seems to argue that Plato's
theory has the result—unwelcome to the latter—that artifact production too will
have to proceed with Forms (e.g., that of Bed) in view.50 Yet it is regularly as-
sumed that Republic 10 in particular introduces such Forms.51 Interpreters are
therefore left with a puzzle, but one that is, I believe, ultimately only apparent.
If one examines those passages in the middle dialogues that appeal to Forms,
one sees that Plato's emphasis on and arguments for their existence involve
multiple types of considerations. A recognition of what one may call, following

49. See 5O4d-5ogd (esp. 5o8b-5ogb), jiyb—c, 5i8c-d. I have in mind here divisions that go
beyond that of genus-species, a prominent example of the latter being the Meno's distinction
between virtue and particular virtues.
50. On Peri Idedn i as a treatment of Plato, see Fine 1993, 34-43. The present discussion is
based on the text and translation of Fine 1993, chap. i.
51. As is often observed, Forms for artifacts are discussed only twice in the middle dialogues
(Cra. 389-390; Rep. 596-598).
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 145

Fine (1993, 57), "the compresence of opposites" in certain entities native to our
world leads Plato to conclude that the reference points in question (e.g., the
natures of justice and courage) must be separate from the spatiotemporal realm.52
When Plato speaks of reference points in the case of artifacts, in contrast, his
claim is not that their existence is necessitated by the compresence of opposites
(beds, in this respect like fingers, are not both F and not-F; see Rep. 59837-10
and Fine 1993,118). Instead, Plato treats them as "organizing principles" (Annas
1976,161) or "functional specifications" (Moravcsik 1992, 66-68) to which prac-
titioners of crafts may appeal as they undertake the process of construction. Of
course, distinct arguments generating different groups of eide might still con-
clude that the entities in question shared, qua ideal, all the same properties. On
the persuasive account of Fine (1993, 86), what Aristotle actually claims in the
Peri Ideon is not that Plato admits no eide in the case of artifacts but rather that
these entities are not supposed by him to share all the characteristics of the core
types of Forms delineated in the Phaedo. Most notably, as she observes (1993,
87), even in Republic 10 there is no suggestion that those eide toward which craft
activity is directed "are everlasting or separate" (cf. Annas 1976, 161).53 Regard-
ing the issue of separation, it is revealing that Plato does not use the locution
auto kath' hauto in his comments on artifact production although in the central
cases of Forms he is at pains to stress just this feature of them (for prominent
occurrences of this phrase, see Phd. j8d6, ioob6; Symp. 2iibi; Rep. 51^5).54
Artifact production is unique insofar as it constitutes the only case in dialogues
up to and including the Republic in which any sort of belief is treated as suffi-
cient for activity involving eide (i.e., there is no other indication in these writ-
ings that doxa is an acceptable cognitive state for techne practitioners on Plato's
strict construction of that type of activity). That this state of affairs would be
nonexistent or at least quite exceptional is entirely to be expected: Plato's point
in introducing the understanding criterion was, after all, to set technai apart from
empeiriai by underscoring that a cognitive state qualitatively higher than that of

52. According to Fine (1993, 61), the middle dialogues evince no clear commitment to sepa-
ration, although "separation fits well with the tenor" of those dialogues. In my view, the middle
dialogues do evince such a commitment. For a defense of this position, see Vlastos (1987), who
maintains that in Plato and in Aristotle's discussion of Plato the same metaphysical claim is
expressed by "The forms exist 'themselves by themselves'" and "The forms exist 'separately'" (187).
53. For further defense of the claim that Plato is sometimes not inclined to invest these enti-
ties with all of the characteristics that Aristotle views as belonging to the central cases of Forms,
see Fine 1993, 86-87.
54. I cannot agree with the claim of Vlastos (1987, 190 n. 11) that Plato's abbreviated expres-
sion at Ora. 386e3 is equivalent in meaning. In my view, Plato's concern in this passage is to stress
that when we make determinations about things, it is not sufficient for us to rely on our private
opinions. As Fine ^993, 86) points out, the entities at issue here "are not said to be perfect or
non-sensible, key features that distinguish Platonic from Socratic forms." What is more, what Plato
says in the Cratylus does not rule out the possibility that shuttle construction could transpire with
undamaged particular shuttles in view (cf. Fine 1993, 293 n. 33).
146 THE ANCIENT QUARREL

belief was required for those activities viewed by him as being of greatest value
and importance. 55 Carpentry as described in Republic 10 thus has a peculiar
status: on the one hand, insofar as it involves Forms, it merits inclusion in the
ranks oftechnai; on the other, the identity of the Forms involved and the cogni-
tive state of carpentry's practitioners explain why, insofar as it belongs to that
group, it has a distinctly lower status vis-a-vis other technai.
The key question for current purposes is whether the practice of poetry, if
revised, could be a techne on a par with carpentry. The main obstacle to its eleva-
tion to techne status might seem to be the fact that, as Plato states on numerous
occasions, poets are divinely inspired, by which he means "possessed" in a very
strong sense of that term. 56 While poets operating on this basis may, and indeed
do, produce beliefs in the audience on the topics of which they speak —notable
examples being virtue and vice —one might think that, in Plato's view, the fact
that poets were inspired or possessed (enthousiastikoi) somehow precluded their
operating with belief as they engaged in the process of creation. If this is so, that
is, if they cannot be assigned a definite position on Plato's belief-knowledge
continuum —even if it is, to begin with, a very unfavorable one—then it is hard
to see how their cognitive state could possibly improve in the way and to the
extent required to permit poetry to be a techne along the lines of carpentry.
In fact, however, Plato's description of poets as passive instruments of the
divine does not constitute the obstacle in question. For one thing, there is a
serious question of whether such talk is to be taken at face value. Woodruff (igSzb,
150 n. 17), for example, maintains that "no plainly nonironical passage tells the
radical inspiration story. . . . Plato never has Socrates speak sincerely of poetic
inspiration . . . no doubt because Plato knew that what Socrates had to say about
inspiration was not strictly true."57 According to Tigerstedt (1969,70), Plato trans-
forms poetical inspiration into possession because endorsing the traditional
model would involve accepting the truth of inspired utterances, thus making it
impossible to subject poets to the requisite control.58 Plato's goal, Tigerstedt
concludes, is to "neutralize the evil effects of the common belief in their inspira-

55. As previously noted, Plato's sharp contrast at Grg. 4540-6 between episteme and pistis
constitutes the first formal appearance in the dialogues of the knowledge-belief distinction. The
contrast is subsequently elaborated, and doxa becomes the standard term for opinion.
56. This view is prominent in the Ion (53^-5353, 535e~536d), Apology (22a-c), Meno (990-
d), and Phaedrus (2453). Regarding Plato's status as innovator in treating the poet as possessed,
frenzied, and passive, see Tigerstedt 1969,1970; Woodruff 1982^ Murray 1992. As Murray observes,
Plato differs most strikingly from his predecessors in his contention that "inspiration is incompat-
ible with techne" -(34; cf. Murray 1981, 99-100).
57. On Plato's lack of commitment to the literal truth of the inspiration story, see also Wood-
ruff 1983, 8-9. Cf. Tigerstedt (1969), whom Woodruff (igSib) too cites in this connection. For a
different view, see Annas (1982, 20), who maintains that when Plato "describes the poet as a di-
vinely inspired being . . . we should take this seriously."
58. As noted, e.g., by Murray (1992, 35), "in the early Greek poets, the divine origin of poetry
is used to guarantee its truth and quality."
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 147

tion, without openly attacking it" (72). Building on Tigerstedt's account, Woodruff
(1982!), 146-47) argues persuasively that Plato has in fact two reasons for telling
"this particular falsehood" about the source of poets' creations: first, "Plato's false
account of poetic experience is a reductio ad absurdum" of claims made for poets'
authority on the grounds that they are inspired; second, "for all its irony, [the
story] has true consequences," especially regarding poets' lack of knowledge of
how to compose and of the subject matter of their compositions.
It is important to recall, in addition, that Plato does not always characterize poets'
state of mind in terms of inspiration. As emphasized in what precedes, both the
Cmtylus and the Republic fault poets for operating with mere belief rather than
with understanding. In fact, as Griswold (1981, 149-50) has noted, "nowhere in
the Republic does Socrates mention the poet's claim to inspiration; indeed, that
claim is pointedly omitted in the passages in which Socrates talks about the begin-
nings of the Iliad (39262-39335)." What is more, Griswold maintains, "Socrates
implicitly denies the validity of that claim throughout the Republic" (150).59 This
state of affairs takes on added significance given the fact that the Republic contains
Plato's most thoroughgoing critical treatment of poetry. It is indeed quite difficult
to see how Plato could conduct his critique, there and in the Cratylus, in the frame-
work of techne—whose requirement on practitioners' cognitive state presupposes
a contrast between belief and knowledge —unless he viewed poets' condition as
describable in terms of this dichotomy.60 It is worth observing, finally, that even in
the Ion—whose image of the chain links rhapsode and poet closely due to their
being divinely inspired —Plato depicts inspiration and a form of belief or reason-
ing as quite compatible.61 Most notably, Ion—who is characterized repeatedly as
speaking as he does about Homer due to divine dispensation—makes clear that
when he speaks he operates based on beliefs he has about the goal toward which
he strives and is able to use means-ends reasoning to bring about the fullest pos-
sible realization of his telos (5356). From the Ion through the Republic, Plato's most
general objection to poets, cognitively speaking, is that they lack knowledge.62

59. On the absence of references to inspiration in the Republic, see also Murray 1992, 39;
Janaway 1995,162. According to Halliwell (1993^ 4), "inspiration never bulks large in, and is indeed
usually altogether absent from, Plato's mature treatments of poetry. There is no trace of it in Rep.
10." For a different view, see Greene (1918, 17), according to whom "Plato at all periods of his life
attributes inspiration to the poets in utter seriousness." Cf. Lodge (1953, 168), who underscores
the centrality of inspiration throughout the corpus ("That the artist is somehow 'inspired,' is
nowhere, even remotely, challenged"); while Lodge provides numerous textual citations in sup-
port of his view (188 nn. 6-8), he does not cite the Republic despite the fact that his book treats
that dialogue extensively.
60. On doxa as Plato's label for the poet's image of reality, see Havelock 1963, chap. 13.
61. For an argument that the Ion views the activities of poet and rhapsode as parallel, one
may consult Murray 1992, 28-29. For poets as the actual target of the dialogue, see Woodruff (1983,
10), according to whom "the danger Plato sees in the rhapsodes lies not in their prestige but in
that of the poets they represent."
62. Cf. Murray 1992, 46.
148 THE ANCIENT QUARREL

Plato's more specific account of their cognitive state focuses alternately on divine
possession and a form of belief, but when he speaks nonironically about their
condition, he concentrates on the latter.
Acceptable poetic activity, unlike the praxis of carpenters, will not involve
eide. It is clear from Plato's discussions in the Phaedo, Symposium, and Repub-
lic that grasping Forms in the realm of values requires nothing short of under-
standing. If poets actually did have knowledge of virtue and vice, hence qualify-
ing as techne practitioners, they would be none other than philosophers.63 Contra
Tate (1928,1932,1938), however, there is reason to think that Plato does not wish
to make poetry and philosophy the task of a single group of practitioners even
under ideal conditions.64 Making philosophers also serve as poets would violate
the Principle of Specialization on which the Republic places so much weight.65
Notably, this principle is invoked in Republic 2 when Socrates says that the state's
founders, who already have distinct responsibilities of their own, are not also to
assume the task of composing stories (3793): "their job is to know the patterns
according to which poets must construct their tales and from which their work
must not be allowed to diverge."66 Here one may draw a contrast between the
activities of naming and poetry: while the former may be placed naturally by
Plato in the purview of the philosopher, since it is in fact an integral part of
philosophical practice, the composition of poetry is a distinct activity that does
not fall in a similarly direct way within the province ofphilosophia. Moreover,
interpreting Plato to hold that there could in principle be a distinct practice of

63. See Phdr. 2y8b-d, where Plato emphasizes that if Homer and other poets had knowledge
and could provide the requisite logos justifying what they had written, the term philosophoi would
properly be applied to them.
64. Several other commentators have attempted to consolidate the two functions. Lodge (1953,
91-92 n. 40, 154), for example, speaks of the philosopher-artist whose poetry reinforces shared
values. According to Greene (1918, 75), Plato "recognized . . . that the poet might express eternal
forms, and so far as he did so, he became a philosopher. In some such way Plato imagined that
the ancient conflict between philosophy and poetry might cease." On this topic, cf. Rutherford
1995, 237 n. 90; Adam 1902, vol. i, 168, and vol. 2, 393. I suggest below that, insofar as the conflict
between philosophy and poetry "ceases," it does so not because the poet becomes a philosopher,
or vice versa. As we will see, in addressing this issue it is crucial that one distinguish between
poetry per se and the use of poetic devices (prominent examples of the latter being myths, simi-
les, analogies, and allegories). For emphasis on the point that the best poetry will not be created
with Forms in view, see also Ferrari 1989, 122; Nehamas 1982, 59-60; Asmis 1992, 358. Regarding
Plotinus' divergence from Plato, see Enneads 5.8.1 and 2.9.16, along with the discussion of Rich
(1960); on Rich's formulation (236), for Plotinus "art will still be, in a sense, a uiuTian; but a uiuTiatQ
that dispenses altogether with a sensible pattern and works straight from the Idea."
65. In using the term "Principle of Specialization," I follow Annas (1981, 73). As introduced
in Republic 2, it is a principle according to which different people have natural aptitudes for dif-
ferent types of jobs, and individuals should pursue those occupations for which they are suited
"by nature" (kata phusin) (3703-0); a refined version of it is the core of Plato's conception of justice
in Republic 4.
66. For a different view of the import of this passage, see Nettleship 1937, 82.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 149

poetry that benefits the state, with its own group of practitioners, fits best with
his repeated emphasis on the special genius of a poet like Homer insofar as such
remarks strongly suggest that an individual on this or a comparable level has a
gift not found in others.67
Although poets are not themselves the practitioners of a techne, they can offer
crucial assistance to philosophers in the performance of their educational func-
tion. Book 10 makes a central contribution to Plato's account of acceptable poetry
by spelling out what cognitive state creators must be in to generate the compo-
sitions judged admissible in Book 3, as well as how this state is situated with
respect both to poets' present condition and to that of those who are to assess
their constructions. In the final book of the Republic, we discover that in order
for poets to create edifying characters in a consistent and compelling manner,
they must have true belief (orthe doxa) about the topics—above all, flourishing,
virtue, and vice —to which their constructions pertain. However, merely having
orthe doxa, regardless of its source, is not sufficient. As in the case of the auxilia-
ries (epikoumi), the crucial distinction regarding poets is between those who have
beliefs, even true ones, which lack the stability of opinions acquired through
paideia, and those having that orthe doxa whose source is education. 68
Cognitively speaking, the latter type of poet is on a par with the carpenter. While
there is common ground between carpentry and poetry from an epistemological
perspective, on the level of objects the activities diverge insofar as the former, but
not the latter, involves eide. In fact, poets' closest counterpart overall is the auxil-
iaries, who also have, not insight, but rather true, stable beliefs about what is and
is not to be feared.69 Auxiliaries' awareness, which does not stem from the appre-
hension of Forms, is nevertheless treated by Plato as an adequate foundation for
the successful performance of their duties. I suggest that, similarly, creators of
acceptable poetry will have true beliefs, inculcated via education, about what one
should and should not grieve over, be angry about, and so on; where emotions
like grief are appropriate, poets will have stable opinions about what constitutes a
moderate response. They will, moreover, have orthe doxa about what constitute
legitimate objects and expressions of desires (e.g., those for food and drink). Poets'
cognitive state will yield reasonably consistent judgments about instances of the
relevant properties (e.g., justice and self-control). As long as someone connected

67. With respect to Homer's talent, see, e.g., Rep. 595b-c, 5g8d, 6066-6073, 6o7c-d; Ion 53ob;
Leg. 68oc. Plato states, moreover, that he praises many things in Homer (Rep. 3833; cf. 3896).
For Plato's use of the term euphuos to underscore the superlative natural capacity that will be
required of acceptable poets, see Rep. 4oic4.
68. See 42gb~43ob for Plato's depiction of the auxiliaries as operating with stable, true belief
about what is and is not to be feared, hence being as consistent as one without knowledge can in
acting courageously. For emphasis on the role of true belief as the epistemological basis of a re-
vised practice of poetry, see also Asmis 1992, 358; Hall ^74.
69. For a recent account of the auxiliaries' values that stresses the stability of their beliefs,
sec Kamtckar rggS.
150 THE ANCIENT QUARREL

to the enterprise has genuine understanding, and hence is able to remove any errors
and inconsistencies from poets' constructions before they assume their final form,
one may be assured that the process will have its desired outcome. The mere fact
of poets' having stable, true beliefs does not, of course, ensure their ability to channel
what is grasped into compelling characterizations. This is where the talent of select
poets comes in. Practitioners of the highest caliber have an aptitude for translat-
ing their beliefs into characters that can take a powerful hold on the imagination.
One might object that Plato's polemic against poetry suggested that it could be of
use to the state only if its practitioners operated with understanding. However, while
Plato emphasizes the necessity of achieving knowledge when the sole alternative
presented for the poet is operating with beliefs that are largely false, or with hap-
hazard combinations of true and false opinions, his view cannot be that only activi-
ties qualifying as technai benefit the state since auxiliaries—who have true belief,
instilled via education, about courage—obviously play a central positive role in
the ideal polls. In the final section of this chapter, I explore more fully the scope
and content of poets' assigned role in Plato's community by discussing his treatment
of early education in Books 2-3, then turning to the account of poetry offered in
the Republic's final book.

6. Poetry's Function in the Ideal Polls

As Plato emphasizes in his treatment of early education in Republic 2-3, one
has a unique opportunity to shape individuals' development when they are young
(for the beginning as pivotal, see 37ya-b, 378d-e, 395d). The influence one has,
which is inevitably momentous, can be negative or positive: everything hinges
on the content and medium of the message. Given poets' status as authorities in
the areas of religion and values, Plato is especially concerned to assess the merit
of their present contributions. Although the material found in poetic creations
is sometimes salutary, often this is not the case.70 In the ideal community, he
insists, the young must not be permitted to hear chance stories constructed by
chance teachers, thus absorbing into their souls opinions (doxai) frequently
opposed to those that they should hold once they reach adulthood (37yb). Plato
is so convinced of the destructive potential of faulty poetic characterizations that
he would rather have children exposed to no poetry at all than allow them to
hear, as in his own time, unexpurgated verses.
Yet, as others have observed, far from rejecting the value to early education
of all poetry, he recognizes its constructive potential and outlines how one might
harness it.71 Plato establishes guidelines that will allow one to determine what is

70. For positive remarks, see 3853, 3896, 3god, 441!), 4680-4693, 5oib, 5i6d.
71. For this acknowledgment of poetry's role, one may consult, e.g., Janaway 1995, 11, 103-4;
Ferrari 1989, 109; Nchamas 1982, 52, and 1988, 214. See also Collingwood (1925, 164), according
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 15!

said well, and what not, in order that the latter type of material might be expunged
and children's souls shaped appropriately by more satisfactory compositions
(37yb-c). Poets are to be faulted above all for their depictions of divinities and
heroes ^yyd-e). Plato is especially concerned to exclude characterizations of
the divine, particularly Zeus, as the source of evil, and depictions of gods and
heroes as overcome by feeling or desire. Acceptable creations, he stresses, will
treat the gods as sources only of the good and characterize gods and heroes as
dismissing or responding moderately to what would ordinarily evoke extreme
emotional and appetitive responses.
The compositions to which the young will be exposed are those that foster
their attraction to what is fine and their repulsion by what is shameful (4016-
4023). Since at this stage reason has yet to develop, and hence cannot be tar-
geted directly, satisfactory poetry helps to create promising conditions for its
optimal emergence: once the right attitudes are formed, reason, as and when it
develops, will reinforce them by enabling individuals to provide logoi justifying
their preferences and aversions.72 Role models have a pivotal influence on the
formation of attitudes and other dimensions of character. In Plato's view, one
models oneself on, thus becoming like, that to which one is drawn. The best
possible "role models" for human beings are the Forms themselves (5<Dob-d).73
Long before one can emulate the Forms, however, one needs virtuous people
to serve as guides. Yet before one can make effective use of human guides, one
is, indeed should be, introduced to admirable figures in the poetry that one begins
to hear at a very young age. The salient point here is that acceptable poetry exposes
people to desirable role models prior to the point at which they can judge reli-
ably between possible candidates in their own physical environment and choose
to emulate those that are genuinely good. If literary influences on children are
positive, at a later stage of their cognitive and affective development individuals
will then gravitate naturally toward — and hence "imitate" — the right sort of
person. With regard to its form, the poetry heard by children will consist not only

to whom "art . . . remains the great educative power by which the young guardians are to be
trained." Plato's description of the three groups in his society, which is sketchy in crucial respects,
devotes little attention to the producer class. The Republic's early curriculum is said to be aimed
at the guardians, who are later subdivided into rulers and auxiliaries. Plato does not address di-
rectly the issue of what portion of this education, if any, children in his third class will receive.
On this topic, see Reeve 1988, 186—91, 309—10 nn. 5—8; Cornford 1941, 63—64, 154 n. i.
72. In my view, the text does not support the assertion of Cooke (1999, 40), for which she
cites no evidence, that once reason develops, "the guardian may change his mind about what is
noble, for better or worse. Thus the early education does not make later criticism impossible.
Criticism is not emphasized early on because the guardian is not capable. But later, when he is
capable, his views of what is kalon are open to change."
73. Plato's conception, which identifies Forms as the paramount objects of emulation, thus
involves the renunciation of a view of ethics as mere role modeling in any familiar sense of
that term.
152 THE ANCIENT QUARREL

of pure narrative but also of narrative effected by mimesis (on permissible objects
of imitation, see 396c-e). Although such mimesis must not dominate poets' crea-
tions, it will nevertheless play a crucial role in children's education by helping
to make good people attractive to the young as role models.74 Under optimal
conditions, poets' genius at the construction and interweaving of the two types
of narrative will lead to the creation of characters that can take a powerful, posi-
tive hold on children's imaginations, with the result that they are drawn to those
figures leading exemplary lives and repelled with equal fervor by their opposites.
The relation between the Republic's early discussion of poetry and Plato's
treatment of it in the dialogue's final book has been, and remains, highly con-
tested. According to several interpreters, there is sharp discontinuity between the
two. Thus, Plato's handling of poetry in Book ro has been described as an "appen-
dix" (Cornford 1941, 321), an "afterthought" (Else 1972, 40), and a "digression"
that is "disconnected from the rest of the Republic" (Nettleship 1937, 340).75 In
the view of Annas (1982), "the split in Plato's thought is radical," such that "any
attempt to harmonize books 3 and 10 must be deeply misconceived" (22, see also
12-13; cf. Annas 1981, 336, 343).76 I hope to show that Plato's presentations are,
not merely consistent, but strongly complementary: with its traditional role as
the core of pedagogy in the forefront of his mind, in Republic 10 he launches a
renewed attack on traditional poetry in light of the metaphysical, psychological,
and epistemological theories of the intervening books.77 As I will argue, more-
over, while Book 10 refines in various respects Plato's earlier critique of poetry,
the scope of what he targets remains the same as it was in Books 2-3. Republic
6073—where he seems to admit only "hymns to the gods and praises of good

74. As Plato emphasizes repeatedly, attitude formation, rather than, say, the assimilation of
information, is the central goal of early education; on this point, cf. Annas 1981, 86.
75. Cf. Adam 1902, vol. 2, 384.
76. See also Urmson (1982, 128-29), wno maintains that "we can represent the argument of
book 10 as resuming that of book 3 only by doing gross violence to one or the other or both."
According to Rowe (1997, 440), in Book 10 "the point that poetry could, ideally, contribute to the
good life . . . is now set aside, in favour of all-out attack."
77. In attributing this level of sophistication to Plato's handling of poetry in Republic 10, I
thus differ from Annas (1981, 335), who describes the book itself as an "excrescence" and main-
tains that the "level of philosophical argument" there falls markedly below that achieved in the
rest of the dialogue. For Book 10 as discussing poetry in light of theories on these topics presented
in Books 4-9, see Nehamas (1988, 214) and Ferrari (1989,120), although my approach differs from
theirs by virtue of its focus on the techne framework. In his earlier discussion of Republic 10,
Nehamas (1982, 51; cf. 53) contends that it fits mainly with Plato's treatment in Books 8-9 of how
the unity of soul and city may be undermined. For Republic 10 as criticizing poets in light of the
ontology and psychology of Republic 4-9, see Kato (1986, 66-69), according to whom, however,
Plato reserves a place for poetry in his hierarchy of technai (70). While I share the view of Ruth-
erford (1995, 211) that Republic 10 deepens Plato's challenge, I cannot agree with his claim (2ir;
cf. 230) that in Republic 2-3 "little more is done than produce a sample anthology of unsatisfac-
tory passages." In developing my argument regarding Plato's critique of poetry in Republic 10, I
have benefited from discussion of the issues with Julius Moravcsik.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 153

people" —is a famous, indeed notorious, stumbling block for those who want to
claim that the two discussions can be harmonized. I contend that 6073, and
Book 10 more generally, concentrates on poetry's civic role, which must be kept
distinct from that pedagogical function previously assigned to it in the context of
early education. On the account to be defended here, Book ro reinforces a posi-
tion that was already indicated by Plato's remarks earlier in the Republic, namely,
that poetry's pedagogical role will be limited to children's education, and poetry
conforming to Plato's strictures about subject matter in Books 2-3 will play a uni-
fying social role on occasions of special importance to the community.78
Plato opens Book 10 by having Socrates affirm that "it is above all the issue of
poetry that convinces me that we were quite right to found our city along the
lines we did." In response to Glaucon's query as to what in particular he has in
mind, Socrates singles out their categorical refusal to admit any "mimetic poetry"
(To u.r|8aia,fj 7capa8e%ea0cu a\>xr\c, oor) IHUIYUKTI, 59535). As Murray (1996,186)
observes, "this statement appears to conflict with the earlier discussion in book 3
. . . where mimetic poetry was not in fact completely banned from the ideal state."
She claims, further (187), that "there is nothing to indicate that the phrase OOTI
pxuT|TiKr| here should be restricted to poetry which imitates the wrong kind of
objects." While I agree with Murray that there is an initial appearance of con-
flict between Books 3 and 10, I cannot endorse her claim about a lack of evi-
dence for this limitation on the scope of what is attacked.
What is noteworthy about the introduction to Republic 10 is that Plato clearly
takes himself to be reaffirming his earlier conclusion (this position is reinforced
by the verbal echo involving 7iapa6e%e(j9ou at 35; cf. 397di and Murray 1996,
186). As Book ro's account of poetry draws to a close, moreover, Plato empha-
sizes (6o7b) that he is reiterating the conclusion sbout whst is to be excluded
that he had previously reached. Given that Plato does not merely begin Book ro
by underscoring this point but repeats it near the end of his second go-round
with poetry—indeed, just after his mention of "hymns to the gods and praises of
good people" (60734) — one should take what he says quite seriously and con-
clude thst he is simply wrong only if there is no way to interpret Republic 10 as
reinforcing what precedes. I suggest that Plato endorses here the same restric-

78. For a different view, see Annas (1982), according to whom Republic 10 indicates that "in
spite of book 3 Plato cannot be considered the advocate of'civic poetry'" (23), and the poetry
admitted at 6073 is "insignificant" (27-28 n. 47). On the view of Nehamas (1982, 69), the poetry
allowed by 6073 seems "negligible and tailor-made for special occasions." While I concur with
the latter portion of Nehamas' remark, I cannot agree with his characterization of what is admit-
ted as trivial. On the connection between 6073 and the civic choruses established in Book 2 of
the Laws, see Ferrari 1989, 141. According to Janaway (1995, 131) in remsrks on Rep. 6073, "Plato
wanders in his description of the poetry he criticizes." On the present interpretation, the target of
Plato's critique remains constant. The shift, insofar as it occurs, is in his identification of what
poetry is to be admitted; even here, however, I would not be inclined to describe Plato as "wan-
dering," since the difference in emphasis is deliberate —reflecting his concern with poetry's dis-
tinct audiences and purposes.
154 THE ANCIENT QUARREL

tions on poetry's content that were described in Book 3, but that he is concerned
now to provide a fuller account of why the poetry in question is not genuinely
beneficial and must therefore be barred from his Republic. As before, Plato's
systematic critique transpires in the framework oftechne. Concerned to dispute
poetry's pedagogical and ethical preeminence, he deepens his investigation of
poetry in light of this framework. Once again, Plato's comments encompass the
goodness, subject-matter, and understanding conditions. This final challenge
to its traditional prominence shows most clearly just how unqualified familiar
poetry is to serve as the source of values and priorities at either the communal or
the individual level.
Book 10 sounds repeatedly the refrain —quite familiar from earlier books of
the Republic and previous dialogues —that traditional poets are panderers par
excellence. Now, however, Plato is more precise about what in us the poet pan-
ders to. Plato's characterization of familiar poetry's destructive impact has more
weight when offered in light of his treatment of the mind in Books 4-9. Based
on his discussion of the soul's tripartite structure and possible tensions among
its elements, we now see clearly that and how, in Plato's view, this poetry strength-
ens what is less worthwhile in us at reason's expense.79 Moreover, if we have
accepted his account of justice as guaranteeing psychic harmony, and the anal-
ogy between justice and health that ends Book 4 (444-445), we will realize that
poetry can make a strong contribution to the establishment of that "diseased"
condition of soul which devalues —and in the limiting case of tyranny renders
valueless —the life on which we are currently embarked. 80 As of Book 9, Plato
has at his disposal, in addition, a distinction between the pleasures of reason and
those of the soul's other aspects (for each element of the soul said to have its

79. Plato reaffirms the existence of intrapsychic tensions at 6c>3c-d. There is controversy over
whether Plato's treatment of the psuche in Book ro clashes significantly with the tripartite con-
ception defended in Book 4. As my remarks in the present chapter make clear, I do not believe
that there is such a conflict.
So. In Book 4, Plato argues that the soul is just when each aspect of it performs its own task
(44id-e; cf. 44309^1). He adds that

having first attained to self-mastery and beautiful order within himself and having harmo-
nized these three principles, the notes or intervals of three terms quite literally the lowest,
the highest, and the mean, and all others there may be between them, and having linked
and bound all three together and made of himself a unit, one person instead of many,
self-controlled and in unison, an individual should then and then only turn to practice if
he find aught to do in either the getting of wealth or the tendance of the body or it may be
in political action or private business —in all such doings, believing and naming the just
and honorable action to be that which preserves and helps to produce this condition of
soul, and wisdom the science that presides over such conduct. (44^4-67; tr. Shorey [1930]
1937, slightly modified)
It is often claimed that Plato equates justice with psychic harmony, but strictly speaking, as Annas
(1981, 132) has noted, "justice is the state that ensures psychic harmony rather than being identi-
cal with it."
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 155

own desires and pleasures, see 58od-58ic). While neither the poetry Plato con-
demns nor that which he favors targets reason directly, in the case of children
the latter type of composition may now be seen to engage what is not inherently
rational, namely, spirit and appetite, in such a way as to facilitate adults' gover-
nance by the pleasures of reason to the extent possible based on the type of nature
possessed by the individuals in question.
As before, Plato explains poetry's current failure to confer benefit in terms of
its shortcomings with respect to subject matter and the cognitive state of its prac-
titioners. Regarding the peri ti condition, he now criticizes poetry in light of his
treatment of Forms and their participants (metechonta) in Books 5-7. Pivotal
here is Plato's identification of Forms as the ultimate reference points of both
judgment and action, and his strong distinction within the realm of appearance
between its inferior and superior constituents. Indeed, one may infer that this
contrast was a prime impetus to Plato's composition of the Republic itself inso-
far as the dialogue defends an account of superlative instances of properties such
as justice (e.g., institutions) that is decidedly at odds with existing valuations.
One's awareness that a contrast within the domain of appearance helps to moti-
vate the Republic in its entirety emerges fully, of course, only after one has read
at least to the end of Book 7. On a more specific level, Plato stresses in Book 5
that aistheta include the conventions of the many about values (^jqd); since
these conventions, along with philosophers' conceptions, are participants in
Forms, the relevant distinction is between metechonta of greater and lesser merit
(i.e., approximation to the truth). In addition, a distinction within the realm of
appearance is made quite explicitly in the Line, where Plato sets the objects of
opinion apart from those of eikasia. From this fine-grained perspective, he is able
to say in Book 10 that familiar poetic creations belong, not simply to the realm
of appearance—which is, after all, inevitable—but that they constitute decid-
edly inferior inhabitants thereof. The analogy with painting (596-598) helps one
to see clearly how low, according to Plato, the ontological status of unrevised
compositions actually is.
Pertinent with respect to the understanding condition is Plato's systematic
distinction between knowledge and belief (474-480) —one that privileges their
differing objects—along with his later refinement of that dichotomy.81 Regard-
ing these subsequent developments, correlated with the Line's distinction within
the realm of appearance is a characterization of superior and inferior vantage
points vis-a-vis inhabitants of that domain according to which the latter perspec-
tive is dubbed eikasia. Eikasia recurs in the Cave, being treated there as the
cognitive state predominant in the populace. As Nettleship (1937, 347) observes,
in Republic 10 Plato draws on his earlier introduction ofeikasia, treating poets-

Si. For a challenge to the standard interpretation of this passage—according to which Plato
endorses the view that knowledge is directed only toward Forms and belief solely toward aistheta —
see Fine 1978.
156 THE ANCIENT QUARREL

who pander to the masses —as operating in that condition. Commenting on

Plato's handling of eikasia in Books 6 and 7, Annas (1981,255-56) points out that
it is not immediately obvious how, or even whether, Plato's accounts of this
cognitive state in the Line and Cave can be reconciled: whereas the former
appears to depict eikasia as a comparative rarity, the latter, which allots it to all
of the Cave's prisoners —who are supposed to represent the unenlightened
masses —treats it as the norm. One attempt at harmonization, mentioned by
Annas (1981, 255), would "admit that eikasia in both Line and Cave means sim-
ply looking at shadows, and . . . urge that what needs expansion is the notion of
looking at shadows." As she observes, however, this solution does not yield a
picture of eikasia as having just the same scope in both cases.
The Line offers a fourfold classification of human cognitive states, the supe-
rior pair of which involve the intelligible realm and the inferior duo the sen-
sible. Plato connects the lowest of the four, eikasia, with objects of a similar caliber
(i.e., "shadows, reflections. . . and so forth," 5103). Yet because the Line's focus
is on the intelligible realm—in particular, on the different orientations toward
Forms represented by the segments of the figure's upper half—Plato does not
elaborate there the connection between eikasia and that toward which its atten-
tion is directed. In contrast to Nettleship, who ties Book ID'S handling of poetry
to the Line's presentation of eikasia, I maintain that the strongest connection is
to Plato's handling of it in the Cave.82 What governs the Line's brief introduc-
tion of eikasia is an emphasis on the inferior status of its objects. The Cave retains
this feature of the Line's presentation but also underscores the fact that the
behavior of those in this cognitive state is utterly indiscriminate: people operating
in this condition simply assume that everything that presents itself to them is,
not only real, but unqualifiedly so. Indeed, a recognition of the existence of a
strong correlation between eikasia and the absence of reflectiveness may be, at
least in part, what leads Plato to broaden its scope in the Cave.
The most salient parallels between the Cave and Book ID'S handling of poetry
lie in the strongly pessimistic attitude expressed toward human beings' typical
level of cognitive activity and in the fact that on both occasions Plato's explicit
concern is with pedagogy.83 Republic 10 offers a distinct refinement of his remarks
in the figure of the Line, aligning itself thereby with the Cave, insofar as it speci-
fies the close relationship between the status of people's attitudes and that of the
objects with which they are concerned. In addition, this final book develops
further Plato's stance in the Cave by applying it to the activity of poets. Thus,

82. Else (1972, 38-39) mentions the Cave in his discussion of Republic 10 but claims that the
latter is tied directly, not to it, but rather to the Sophist. Else takes this as one source of support
for his contention (56) that the treatment of poetry in Republic 10 postdates the Sophist.
83. Notably, Plato's introduction to the Cave (5143) stresses that his paramount concern is
with the effects of proper education and its absence; in Book ro, see 59gd, 6ooc, 6o6e. Plato's
judgment of the poet's audience (i.e., the populace) is quite harsh, of course, not only in the Cave
and Book 10 but also in his earlier description (Book 5) of the lovers of sights and sounds.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 157

Plato leads into his treatment of poetic mimesis by observing that "mimetic activity
is capable of producing everything because it has only slight contact with each
thing, and what it has access to is a mere simulacrum" (5g8b).84 Moreover, while
the Cave presents shadows and those entities whose shadows they are as being
accessible at distinct stages of cognitive activity (jijc-d), in Book 10 Plato stresses
their contemporaneous presence by having Socrates clarify that "I am speaking,
not about anything difficult, but rather about a state of affairs that can be easily
generated. The fastest way of doing so, I suppose, would be to get ahold of a
mirror and carry it with you everywhere. You will then quickly find yourself able
to make all the things just mentioned, including the sun and heavenly bodies,
the earth, yourself, other animals, plants, and inanimate objects" (5g6d-e). These
remarks, which help to set up Plato's discussion of poetic activity, convey most
vividly of all the indiscriminate character of the activity targeted by him. Finally,
while Plato's handling of the dynamic of slavery and liberation in the Cave might
be taken to suggest that people are somehow constrained by forces beyond them
to take what are merely shadows to constitute reality, his talk of the mirror in
Book 10 is noteworthy insofar as it underscores his view that the lack of selectiv-
ity involved is by no means wholly involuntary.
Plato might have made —and elsewhere does offer —his criticism of familiar
poets in terms of belief, namely, by speaking, not of eikasia, but rather of ran-
dom combinations of false opinions with those that happen to be true. In the
present context, however, Plato's use of a term other than doxa to characterize
the poet's level of cognition —which is in keeping with the overall tone and
purpose of the discussion —allows him to foreground the qualitative difference
between this creator's mental condition, on the one hand, and that of auxilia-
ries, along with Book 3'$ "austere" poet (^gSa-b), on the other. Most significantly,
Plato's terminology reflects his view that the cognitive state of those who dispense
and accept indiscriminately misconceptions about the matters of greatest im-
port is qualitatively different from that of individuals who have true beliefs, in-
stilled via education, about these topics. If there is such a stark contrast between
these states, then, strictly speaking, they should have different onomata.
As I have suggested, compared with that in Books 2-3, Plato's critique of poetry
in Book 10 is deeper in key respects. The outcome of this further treatment of
poetry via the techne framework is that Plato has an even stronger basis than before
for his denial of its entitlement to this status and hence for his rejection of its
claim to ethical and pedagogical authority. A sharpened challenge need not,
however, issue in an enlargement of one's critical focus. In contrast to Murray
(1996), I maintain that in Book 10 Plato in fact targets the same poetry that he
had previously rejected. Book 3 admits the restrained poet and excludes the one
who panders: although the latter is far more pleasing, the former is the only one
capable of benefiting individuals and the community (3983—b). While Book 10

84. I adopt here Halliwell's rendering of eidolon as "mere simulacrum."

158 THE ANCIENT QUARREL

may seem initially to attack additional poetry, Plato stresses that he is targeting
once again the practitioner who acts in an indiscriminate manner due to his
paramount concern with catering to whatever preferences and aversions his
audience, namely, the populace, happens to have.
Those who maintain that what Republic 10 targets is broader in scope opt for
an interpretation of mimesis according to which now, in contrast to Book 3, all
poetry involving it is encompassed by Plato's critique. In my view, his construc-
tions of mimesis in the two books are quite compatible in the sense of leading to
the exclusion of the same class of compositions.85 Tate (1928,1932) attempts to
reconcile Plato's treatments of mimesis in Books 3 and 10 by attributing to him
a distinction between good and bad senses, or kinds, of imitation. With Nehamas
(1982, 49-50), I maintain, in contrast, that the crucial distinction for Plato in-
volves the objects of activity. Although Book 10 seems initially to dismiss all poetry
incorporating mimesis, I suggest that Plato's subsequent emphasis on pleasure
(605-607) casts doubt on this conclusion.86 In fact, in a noteworthy passage, Plato
provides an extended formulation (60704-5) of what he wishes to exclude, of
which his phrase at 5953 may be viewed as a compressed version: Halliwell, who
rightly identifies the construction as involving a hendiadys (1993^ ad loc.), ren-
ders the expression as "poetic mimesis designed for pleasure" (fj rcpoq T|6ovfiv
7ionyuKf| Kod fi (jliiTicic;).87 This remark is complemented by Plato's indication
at 6046 that in Book 10—as before—not all mimesis is encompassed by his at-
tack: in this passage Plato distinguishes the typical objects of poetic depictions,
which are his target, from representations of figures whose conduct is reason-
able and hence moderate; mimesis of the latter objects, while dubbed more
challenging, is nevertheless treated as both desirable and possible. These com-
ments, taken together, support the present interpretation, according to which
the nonpandering mimesis of good objects permitted in Book 3 is not included
as a target of Plato's subsequent challenge.
One must consider, however, whether—as has often been thought—the view
that Republic 2-3 and 10 reach the same conclusion about what is to be exclude
is challenged by Plato's famous insistence near the end of the latter discussion
that he will admit into the ideal city only "hymns to the gods and praises of good
people" .88 It is indisputable

85. For this stance, see also Asmis 1992, 350; Ferrari 1989,125; Belfiore 1984,126-28. Cf. Janaway
1995,126: "The sense of the term mimesis changes between Book 3 and Book 10 without losing its
reference to the same kinds of poetry."
86. In this lengthy passage, Plato makes clear use of Book g's insistence on the existence of
pleasures of reason and their fundamentally superior caliber.
87. For an earlier instance of hendiadys in Book 10, see 595bg and Halliwell ig93b, ad loc.
88. As an explanation of the fact that only in Republic 10 does Plato turn explicitly to the
question of acceptable poetry that depicts human beings—as distinct from divinities and heroes —
one may wish to invoke his earlier comment (392C1-4) that one can specify what is permitted in
their case only after the discussion has revealed the nature of justice and demonstrated its benefit
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 159

that, according to Plato, the compositions favored here constitute a decidedly

higher subset of entities in the domain of appearance —that is, they instantiate
more fully and reliably the relevant properties, notably "fineness" (to kalon) —
than those generated by existing poets.89 Instead of praising whatever happens
to gratify the populace, these compositions will laud only figures and actions
that are genuinely worthy of this treatment.
Plato's remark at 6073 does indeed present a difficulty if he is interpreted
to claim that the only poetic compositions now acceptable to him are those
belonging to the genres of hymn and encomium as typically construed with
regard to their form. One way of challenging this interpretation of the passage
would be to argue that one need not regard Plato as subscribing therein to
narrow generic constraints on the form of acceptable compositions, that one
may in fact treat the passage as itself repeating the view of acceptable poetry
expressed elsewhere in the Republic. From this perspective, it would be impor-
tant to keep in mind that the poetry admitted in Book 3 —constructed in the
"unmixed" style that represents those who are good (39yd) —will not be famil-
iar in the sense of fitting neatly into any existing genre. To cite Plato's favorite
example, namely, Homer, while he admits that there is good to be found in
this poet's creations, the compositions described by Plato would not simply be
revised epics. The very fact that the Republic so fundamentally criticizes exist-
ing generic boundaries, and does not hesitate to violate traditional distinctions
in offering its own recommendations, might make one reluctant to assume that

to those who possess this quality regardless of whether they appear to have it. On this issue, see
Gould (1990, 29 n. i) and Reeve (1988, 222), according to whom "the imitative poetry under dis-
cussion in Book 10 is about human beings and their actions or activities"; having quoted Plato's
remarks at 60304-9, Reeve concludes (222) that "the discussion of poetry in Book 10 fills the la-
cuna in the earlier discussion, and occurs at the logically appropriate point in the overall argu-
ment." For the denial of a connection to 3920, see Adam 1902, vol. 2, 384.
89. On this characteristic of superior poetic compositions, see Plato's earlier claim that one
must seek creators whose natures qualify them to manifest beauty (to kalon) and grace in their
work (4orc). He wants the influence of fine works (ta kala erga) to be absorbed, thus guiding the
young without their notice to likeness and harmony toi kaloi logoi (c-d); in his view, in fact, the
culmination of mousike is ta tou kalou erotika (4030). The best condition for a human being is
being admirable (kalon) in soul and body (4O2d), especially the former; cf. his later comment
that virtue itself is a kind of beauty of soul (hallos psuches, 444d-e). As Book 3 makes clear, "fine"
poetry helps one's soul to move in the direction of acquiring this beauty. Although Book 5*5 lovers
of sights and sounds —who run from one dramatic festival to another—are concerned in some
sense with beauty, they are typically absorbed by quite inferior manifestations thereof; cf. Plato's
comment in Book ro (6cab) that the poetic imitator will depict what seems beautiful to the igno-
rant multitude. In Book 6, Plato describes the Form of the Good as more beautiful (kallion) than
knowledge and truth (jo8e). It thus turns out that though both philosophers and the lovers of
sights and sounds are occupied with beauty, the objects of their ultimate concern are fundamen-
tally different in ontological status and, therefore, value. As indicated above, in the ideal com-
munity, poets' constructions will no longer reinforce people's occupation with wholly inadequate
manifestations of this crucial property.
16O THE ANCIENT QUARREL

only a narrow (i.e., conventional) interpretation of a term like "encomium" is

plausible.
Regarding Platonic methodology, Andrea Nightingale's discussion (1995,
chap. 3) of Plato's critique of the prose encomium is instructive. According to
Nightingale (93), while Plato is interested in rhetoric as a whole, he is especially
worried about rhetoric of praise. Notably, she emphasizes the importance to Plato
of "the discourse of praise that is not packaged in a formal speech or composi-
tion" (104; more generally, 104-6); as she stresses (108), his concern is with its
pedagogical and ethical import, not its form (i.e., aesthetic status). One might
be tempted to make the same point with regard to poetry, observing, for example,
that in the Republic Plato's occupation with informal provisions is evidenced by
his keen interest in challenging that extravagant praise of Achilles embedded in
the Iliad.90 In fact, Plato's target in the dialogues seems to be quite general,
encompassing, to borrow Nightingale's phrase, all "ignorant conferrals of praise"
(1995,120)—whether they be formal or informal, offered in prose or in verse.91
If the encomia criticized by Plato are so multifarious, would one expect him to
revert to a narrow focus when making his own recommendations, especially when
his own discussion in Book 3 tells against this view, and when he states twice in
Book 10 that he is simply reaffirming his earlier position? While this may not be
the outcome one anticipates, I suggest that this is in fact what happens at 6073.
In the dialogues, Plato uses the verb encomiazo in broad and narrow senses.
In the Republic, it appears on numerous occasions, being employed both where
he criticizes the misguided use of language of praise and where his focus is on
what it is acceptable or desirable to laud.92 The verb's looser meaning is in the
foreground, as when Plato employs it in remarks that bear on the dialogue's aim
of lauding justice (358d2, 363d5, 36yd7, 589b8); in critical observations on the
transition from an "oligarchic" to a "democratic" psychic constitution (56064);

go. For strong emphasis placed on the status of Achilles, see Murray 1996, 156: "Part of P.'s
purpose in constructing his educational programme is to replace Achilles as a role model for the
young with that of the philosopher as the highest human type."
91. Nightingale uses this phrase (in the singular) when commenting on Alcibiades' encomium
of Socrates, which she interprets as a "caveat" against the bestowal of praise in ignorance of the
true character of that toward which one's eulogy is directed. According to Nightingale, Plato's
multiple critiques of encomiastic discourse have two central aims: "First, to show how the rheto-
ric of praise that pervades classical Athens can damage individuals and, indeed, the city as a whole.
And second, to demonstrate the antithesis between praise discourse and the language of the phi-
losopher" (131). In fact, those features identified by Nightingale as lying at the core of Plato's cri-
tique of encomiastic rhetoric—whether formal or informal—are the same as those offered in the
Republic with regard to poetic compositions: What is praised may be in reality good (e.g., 39yd)
or bad (e.g., 568b). Where the objects are bad, the conferral of praise signals its provider's lack of
awareness and causes harm both to individuals and to the community. Notably, in Book 10, Plato
rejects poets' familiar provisions of praise as not issuing from those in a position to confer it
reflectively (jggb). Philosophers, in contrast to poets, will dispense praise and blame as these are
merited by the objects of their attention.
92. See 358d2, 363d5, 36yd7, 56064, 568b3, 581010, 583d8, 58gb8, 5ggby.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY l6l

when observing that those in pain praise the mere freedom from this condition
as the highest pleasure (583d8); and to make the point that individuals leading
the three types of lives he delineates —namely, those devoted to wisdom, honor,
and gain—all commend their mode of existence as the most pleasurable (581010).
The noun encomion appears only once in the Republic, at 60734. While Plato's
use of the verb in the dialogue certainly does not rule out the possibility that the
noun at 6oya is to be interpreted along similar lines, making a direct inference
from the prominence in the Republic of the verb's broader meaning to the noun's
meaning at 6073 would be problematic.
Reflection on occurrences of encomion outside the Republic shows that it is
used consistently along standard lines (see, e.g., Ion 53403; Prf. 32632; Symp. lyybi,
I94d6; Leg. 82zb5,7,95869). On one noteworthy occasion, however, following his
remarks in praise of ems in the Symposium—which depart substantially from the
familiar style—Socrates tells Phaedrus that "you may call this speech my encomium
of ems, if you like; if you would prefer not to use this term, call it whatever you
wish" (tomov... TOV Xoyov... ei tiev (knjAei, ox; eyKcb|iiov ei<; "Epcora VOLHOOV
etcrno9oa, ei 8e, OTI KCU onr\ %aipei<; ovoua^cflv, tomo ovoiia^e, 2i2b8-c3). This
comment indicates that the form of a composition is quite salient, at least from a
certain perspective, when it comes to answering the question of whether the term
encomion is properly applied to it. What cannot be determined is whether Plato's
inclusion of Socrates' speech alongside the preceding compositions, combined
with his observation at 2iib8-c3, signals his own willingness to apply this term to
constructions that fall outside the traditional range yet meet his strict criteria with
respect to content. In the absence of further evidence, one therefore lacks a basis
for concluding with any assurance that Plato intends to include compositions
departing significantly from the traditional form among those encomia mentioned
at Republic 6073.
This impression of Plato's focus in 6073 is confirmed when we turn to the
case of humnoi. In the dialogues, Plato sometimes employs the verb humneo
with a broader meaning (see, e.g., Prf. 31736, 343b3; Rep. 36431,463d7; Tht. 17631;
Tim. 47b4; Griff. n8b3). This is not the case, however, with the noun humnos.
In addition, one finds no passage that plays a role for the term humnos thst is
comparable, with respect to its possible suggestiveness, to Symposium 2i2b-c in
the case of encomia. Moreover, and quite significantly, in the Republic itself the
noun humnos is found twice aside from 60734, on both occasions with what is
clearly a narrow meaning (45906, 468dg). Such evidence, combined with the
fact that Plato's expression at 60734 singles out the gods 35 the focus of the com-
positions in question, strongly supports 3 conventional interpretation of the form
of the humnoi mentioned there.93 The presence of kai in a coordinating role

93- Notably, in Republic 2 he uses the verb humneo with theoi as the direct object ^yiby-S)
in a discussion of the "city of pigs" in which compositions that are traditional with respect to their
form are unmistakably at issue.
162 THE ANCIENT QUARREL

(VLIVOW; Geoii; Km syKcoiua ~colc, cryaGoli;) provides a clear indication that the
terms humnoi and encomia are to be interpreted along the same lines with regard
to their scope of application. Thus, even if one wanted to leave open the possi-
bility, following an investigation of encomia, that it admitted a broader construc-
tion at 6073, the fact that this option is not available in the case of humnoi would
preclude one's offering such an interpretation of Plato's expression.
If the encomia and humnoi at issue in 6073 are to be construed along familiar
generic lines, then the most defensible position is that the compositions referred
to there are not the same, with regard to their form, as those singled out as accept-
able in Book 3. It need not follow, however, that Plato is being inconsistent or that
his position has changed after all such that his condemnation now encompasses
more poetry than before. One may, of course, wish to opt for one of these posi-
tions. Another possibility, however, is that his comment about admissible poetry
is made with a different audience and purpose in mind. Since, as we have seen,
Plato insists in Book 10 that the poetry targeted by his critique remains the same,
and one finds good evidence there to support that view, it seems reasonable to seek
an interpretation of 6073 that is not at odds with this position. Only if such a con-
struction turns out to be indefensible should one conclude that Plato's discussion
incorporates a flaw on the level that would be at issue if 6073 clashed in such 3
striking and serious way with his claims elsewhere in Book 10. In what follows, I
offer support to the view that Plato's focus in Book 10 is on adults and on poetry's
potential role, not in pedagogy—as it was in his reflections on children in Books 2-
3 —but instead in the fostering of civic unity.94 On the view presented here, it thus
turns out that one who wishes to challenge the coherence of the Republic's han-
dling of poetry cannot place a heavy reliance on 6073 to accomplish that task.
Plato's discussion of eponymy in the Phaedo and treatments of the Method
of Division (diairesis) in the Sophist and Politicus leave no doubt that naming,
insofar as it involves the connection between onomata and nstures, will be ac-
complished by the philosopher. Under ideal circumstances, poets, supervised
by philosophers, will utilize that inquiry's results. Poets will be prohibited from
exploiting the resources of language to promote undesirable attitudes and emo-
tional responses or based on the wrong conception of phusis. They will, how-
ever, be actively encouraged to use these resources to produce the correct set of
attitudes and responses via poems that are composites of simple narration and
mimesis, above all of good men acting "steadfastly and intelligently" (aGfyokotc,
TE Km eu<j>p6va><;, 396di). In this ideal collaborative setting, poets will construct
onomata — especially proper names of outstanding characters—with positive
descriptive content, incorporating them in their tales to the extent that philoso-
phers judge their so doing to be helpful in educating the young.95 That Plato

94. For Republic 10 as concentrating on adults, cf. Nehamas 1982.

95. That is, insofar as the inclusion of such onomata—whose semantic breakdown would need
to be readily apparent—is thought to contribute in particular cases to the ultimate goal of focusing
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 163

envisions such a role for these authors and onomata is evident in Republic 3,
where he rejects the use of names whose negative descriptive content is designed
to instill incapacitating fear but insists that names of the opposite type be
employed in speech and poetry (3870). More generally, poets' education, supple-
mented by guidelines received from the philosopher, will aid them in produc-
ing the desired result (namely, vivid and engaging portraits of suitable role
models) with greater reliability, and under the philosopher's watchful eye, any
remaining infelicities can be excised before the poetry reaches youthful ears.
Under this description, poets —qua early educators—thus serve as "auxiliaries"
of a kind, in the unique way permitted by their gift, by helping philosopher-rulers
to achieve and sustain the flourishing of the polis.%
In contrast, as Republic j's treatment of the advanced curriculum makes clear,
when it comes to individuals' transition to understanding Forms, and to the
solidification of that understanding, poetry proper is not involved. As Plato
observes (498^, while they are children, individuals should pursue an educa-
tion and culture appropriate to youth. As their souls approach maturity, how-
ever, the exercises assigned to people should be altered to reflect their new cog-
nitive state and capabilities. Already in Book 6, Plato indicates the necessity of
considering paideia again (5026-5033), this time for the purpose of describing
those areas of inquiry that will constitute the advanced curriculum. The exis-
tence of a break in continuity regarding the content of the two curricula is under-
scored in Book 7 (522a2-bi), where Plato stresses that the earlier studies, in con-
trast to those with which he is now concerned, do not generate episteme. The
question now is, which studies do promote and issue in understanding? In what
follows (522b-54ib), Plato identifies and describes the relevant activities, and the
general time frame within which each should be pursued. These studies, in
contrast to early paideia, target reason directly (527d-e, 5320; cf. 534b); as such,
they complement the latter, which helps to pave the way for them. In sharp
contrast to those comparatively abbreviated time periods familiar to us, Plato's
advanced course of study takes individuals well into adulthood—up to the age
of fifty or so—at which point a select few apprehend the Good (5403).
As Plato's own practice in the Republic illustrates, instead of resorting to
poetry per se to promote and solidify understanding, philosophers will employ
literary devices such as myths and allegories where relevant. Unlike poems,

individuals' attention on the relevant abstract entities. Poets must not, of course, entertain or convey
any illusion that their constructions reveal individual natures, whose status as fundamental is
rejected in the Cratylus. As previously emphasized, all determinations of correctness involving
primary entities and their onomata will be made by the philosopher.
96. One might wonder whether this function, in theory open to poetry, could be assumed by it
in practice. In the view of Annas (1982, 23), a talent stemming from the divine could not be har-
nessed to Plato's pedagogical ends. If, as emphasized in what precedes, Plato's talk of poets' inspira-
tion is not to be taken at face value, then this sort of objection loses at least some of its force.
164 THE ANCIENT QUARREL

these must be constructed by philosophers themselves since they presuppose

and complement insights presented through rational argumentation. Hence,
in this case, unlike that of poetry proper, practitioners must operate with episteme
rather than orthe doxa. That dianoia too, as described in the Line, will be
inadequate for this task is evident from Plato's emphasis on the directness of
apprehension that sets philosophers' cognitive state apart from that of other
techne practitioners.
On the foregoing interpretation, Plato retains a place for the poetic in two
distinct senses: one makes it the unique province of the philosopher; the other,
of the poet who is assisted by the philosopher. Given this twofold use of the
concept of the poetic, solutions to the problem of Plato's attitude toward poetry
that center on talk of the "poet-philosopher" are potentially misleading if intended
to suggest that, in the ideal polls, all that may be called "poetic" must be accom-
plished by philosophy. As I have argued, this is not in fact the case.
Philosophers themselves will have no need of poetry. Yet one might think
that, based on the parameters of his own thinking, Plato should allocate a
pedagogical role to it in the lives of nonphilosopher adults, in particular those
belonging to the producer class. He does not do this, however. What he does
instead is make clear at various points in the Republic that poetry whose content
is positive is one of many tools that philosopher-rulers may use to reinforce civic
unity (see 372^ 4596-4603, 468d; cf. 465d-e and 54ob-c).97 Thus, in articulat-
ing Plato's view of poetry in the dialogue as a whole, one must distinguish care-
fully between compositions (e.g., praises of heroic conduct) intended for the
community at large and constructions designed for pedagogical purposes —
namely, to promote the formation of the right attitudes and attraction to good
role models—which would be tailored specifically to children and heard exclu-
sively by them.98
While Books 2-3 focus on children, in whom the capacity to reason has
not yet developed, there are clear indications that Book 10 concentrates on
adults. At 6o2C, Plato asks, with respect to which aspect of the soul does poetry

97. As the interpretation of 6073 offered here indicates, I cannot agree with Else (1972, 52
n. 79), according to whom Plato does not give a clear indication in the Republic of the context in
which the hymns and encomia of 60734 will be relevant; notably, Else's own references to Pla-
tonic examples from the two genres (52—53) do not include any of the passages cited here. For
mention of the first three passages cited above in connection with remarks on 6073, see also
Halliwell rgg3b, rj3; Murray 1996, 229. On the interpretation of Murray (^996, 185), which dif-
fers substantially from my own, in Republic 10 Plato's hostility to poetry is "treated in a far more
trenchant manner than in the earlier books. There P. was concerned to reform existing poetry
through censorship, and was primarily interested in the effects of poetry on the young, whereas
now his aim is to remove poetry from Greek culture altogether (60733-5)"; cf. Murphy ^67, 224.
98. While I am in accord with the claim of White (1979) that Rep. 6073 and the dialogue's
earlier discussion do not clash with respect to the identity of the poetry that the ideal state will
admit, I cannot agree that Plato's focus st the two junctures is on precisely the same class of com-
positions (257, 259).
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY 165

have its power (dunamis)? In what follows (6020-6033), he draws a sharp con-
trast between its rational and inferior aspects. It is obvious, he maintains, that
familiar poetry targets and strengthens what is inherently irrational at reason's
expense (6o4d-6o5c; for a general statement regarding poetry's cultivation of
feeling and desire, see 6o6d).99 That adults are his focus at this juncture is
evident from his concern to stress that exposure to poetry with problematic
content typically results in the degeneration of the minds (lobe tes dianoias)
of the individuals in question (5953-^); this degeneration consists in reason's
being overcome by the irrational aspects of the soul, namely, emotion and
appetite (6o6c-d).100 Plato discusses at length (6o3e-6o4d) the case of the
grieving man who has experienced the loss of someone or something dear to
him, for example, his son. On an occasion of this sort, Plato says, reason
encourages him to resist capitulating to his grief, while the feeling itself (auto
to pathos) struggles to take control (6o4aio-bi). Giving way to grief in such
instances prevents us from making effective use of our reason to determine
just what has happened and where we should go from there, encouraging us
instead to conduct ourselves like children (6o4c; on the topic of grief, see also
6o5c-6o6b, and regarding the laughable, see 6o6c). Notably, Plato identifies
as his gravest concern the possibility that even superior individusls, nsmely,
those in whom reason slready has a fairly strong presence —but not a solid or
fully developed one, as in the case of philosopher-rulers —will be conquered
in this way (6o5c). 101 This emphasis on the degeneration of adults' souls pro-
vides one with essential guidance regsrding the context in which to interpret
his remark about acceptable forms of poetic composition at 6073, which sppears
to limit poetry to a narrowly circumscribed civic role. Plato's prescription there
does not supersede that offered in Book 3 (3960-6). Rather, it supplements what
he says earlier by stating explicitly what was already apparent, with respect to
those who are approaching or have reached maturity, in the intervening books.
As evidence of the strong continuity between Republic 10 and what transpires
earlier in the dialogue, one may point to the notable omission from the advanced
curriculum of poetry per se, combined with the fact that Plato's examples prior
to Book 10 of viable compositions to which the polls at large will be exposed
fall quite nicely under the rubric of 6073.
With regard to the structure of the Republic's bipartite discussion of poetry, I
thus venture the following conclusions. The most central criteria that Plato offers
for the content of viable poetry in Books 2-3 are that it must depict the gods as

99. In articulating his position on this issue, Plato refers back explicitly to the Principle of
Conflict, of which he had made heavy use in Republic 4 when arguing for the tripartite structure
of the soul (6c>3d; see also 604!), which is strongly reminiscent of 436b8-ci). The phrase "Prin-
ciple of Conflict" is drawn from Annas 1981, 137.
100. Regarding 5953-!), cf. Nehamas 1982, 72 n. 19.
101. On the existence of exceptional cases, cf. 60605-7.
166 THE ANCIENT QUARREL

responsible solely for the good and that mortals and divinities who are to serve
as exemplars for budding and full-fledged members of the community must not
be depicted as subject to strong desire and emotion. Plato's point here is that
any poetry that fails to conform to these restrictions will be excluded straight-
away from his ideal community (on the general scope of his strictures regarding
the content of poetry, see 3y8c-d, 383^ 3863, 387^. In Books 2-3, Plato spells
out what follows from the existence of these constraints with respect to poetry in
its pedagogical role. This early discussion does not address directly the question
of what other function, if any, poetry might have in his community. In Book 10,
Plato treats this topic explicitly, underscoring the fact that the same restrictions
on poetry's subject matter are in place and considering what follows from their
existence with regard to the broader civic role that poetry may perform. Since
Plato has already established in Books 2-3 the necessary conditions that poetry
must meet with regard to its content in order to be eligible for inclusion in the
state in any capacity whatsoever, and has reiterated in Book 10 his adherence to
the position articulated there about what will not make the grade, what remains
is for him to state the form of those compositions that are to play a role in the
marking of important civic occasions, notable among them ceremonies honor-
ing those who acquit themselves well in battle and those who have governed the
community with integrity and effectiveness.
Undeniably, Plato's insistence that poets may serve only as elementary
educators, and even then only under ideal conditions, constitutes a significant
demotion insofar as, according to tradition, they were the instructors of adults
(Aristophanes, Frogs 1054-55).102 On Plato's model, as developed above all in
the Republic, it is philosophers, not poets, who play the latter role. In this way,
and by functioning as the community's ultimate authority regarding the con-
tent and form of admissible constructions, philosophy clearly supersedes poetry
as the educator of Greece. Poetry's role, however, while clearly circumscribed,
is far from negligible since, as Plato himself notes repeatedly, early influences
on the young make a —even the—crucial difference to the kinds of adults they
become, in particular, to the way in and extent to which they realize their natures.
Moreover, as we saw, the community at large will be the audience for poetic

102. There is disagreement over how tois men gar paidarioisin esti didaskalos hostis phrazei
should be interpreted. Dover (1993, 324) favors '"boys have a teacher who explains (things to them)'"
over "'anyone who speaks to boys is (ipso facto) their teacher.'" Stanford (1963,165) and Lattimore
(1969, 68) also prefer the former interpretation, but Murray (1916, 78) endorses the latter. The
outcome of this debate is not relevant to the project at hand since, on either interpretation, the
point about the stage at which poets make their contribution remains the same. For an impas-
sioned defense of the salience ef art, particularly literature, that is prompted by Plato's criticisms,
see Murdoch 1977, 76-89; with respect to pedagogy, she contends (86) that "art is far and away
the most educational thing we have, far more so than its rivals, philosophy and theology and sci-
ence." For positive remarks, directed against Plato, on the value of literature for adults, see also
Gould 1990.
 THE QUARREL BETWEEN PHILOSOPHY AND POETRY l6/

constructions that help to mark important civic occasions. The foregoing account
of the Cratylus and Republic serves as a powerful reminder of how central the
quarrel with poetry is for Plato and how determined he is to triumph. At the same
time, this discussion —conjoined with the investigation of the poetic backdrop
for Plato's own theories about orthotes onomaton — underscores the fact that he
does not wish to do so at any cost, above all at the price of denying the ideal polls
the contributions of a potential benefactor.
 CONCLUSION

Plato's ultimate stance toward poetry, or art generally, has often been seen as
extraordinarily harsh. Thus Schaper (1968, 54) writes that he "had to condemn
the whole of poetry because it claimed the very status he was trying to deny it";
if in fact exposure to poets generates "the kind of frenzy and unreflective identi-
fication which Plato outlined, then they must go. ... To measure art by the stan-
dard of knowledge — or, for that matter, by the standards of anything else — is
always to find it wanting." In the view of Murphy (1967,224), Republic 10 "deals
especially with the claim of art to teach and to impose ideals, but it cannot fairly
be said that Plato would have tolerated it in his chosen city if the claim were
renounced and it had remained in its own province." According to Murdoch
(1977,43), in turn, for Plato "art is pleasure-seeking self-satisfied pseudo-analysis
and pseudo-enlightenment. . . . Nature educates us, art does not." In Plato's view,
Murdoch continues (1977, 65), it "is playful in a sinister sense, full of. . . a spite-
ful amused acceptance of evil, and through buffoonery and mockery weakens
moral discrimination. . . . Art is sophistry, at best an ironic mimesis whose fake
'truthfulness' is a subtle enemy of virtue."
If interpreted along these lines, Plato comes to be seen as the source in West-
ern thought of a dichotomy between philosophy and literature that is at best
undesirable and at worst highly pernicious. In the opinion of Murdoch (1977,
81-82), the "ancient quarrel" that was Plato's legacy to the West has long en-
dured, such that only "comparatively recently" have philosophers turned to litera-
ture in ways that evince a serious recognition of its pertinence to their concerns.1
On Danto's formulation in the essay "The Philosophical Disenfranchisement of
Art," in which he comments on the Republic's handling of literature, Plato wishes
us to "accept a picture of the world in which the place of art is outside it. And

i. Murdoch's own focus here is on moral philosophy.

168
 CONCLUSION 169

since Plato's theory of art is his philosophy, and since philosophy down the ages
has consisted in placing codicils to the platonic testament, philosophy itself may
just be the disenfranchisement of art" (1986, j ) . 2 Danto's own avowed aim (1986,
xv) is to undertake the "reenfranchisement" of art against Plato's strictures. More
recently still, Edmundson (1995), who cites Danto's interpretation of Plato ap-
provingly (7), contends that "literary criticism in the West begins with the wish
that literature disappear. . . . For to Plato poetry is a deception. . . . When he
conceives his Utopia, Plato banishes the poet outside its walls. . . . Is there any
other kind of intellectual inquiry that originates in a wish to do away with its
object?" (i). Returning to this issue a little later, Edmundson asserts that literary
criticism starts "with the conviction, expressed by the greatest of philosophers
. . . that poetry is a harmful diversion, best repudiated in the self and cast from
the state. Effective literary criticism ever after attempts to defend poetry against
this heaping Platonic insult" (7).
Although one may not agree with the specifics of Plato's account regarding the
ties between philosophy and literature, his view nonetheless represents a substan-
tial achievement. While Plato is critical of the Greek literary tradition, he pays it
the tribute of taking it, where relevant, as an important locus of ideas. I have ar-
gued in what precedes that in order to understand the unfolding of Plato's approach
to key issues in the philosophy of language, one must view it against the backdrop
of literary activity. Plato's own ideas about reality and language influence his reac-
tions to his literary sources. Upon reflection, Plato concludes that linguistic de-
vices prominent in literature may be employed in two ways: Certain ones can be
used to advantage in the conduct of philosophy. Where a literary device is not
employable in philosophy, Plato believes that it is nevertheless instructive to indi-
cate why one may not justifiably invoke it in that context.
As we saw in the investigation of the Cratylus in chapter 2, earlier commen-
tators have been reluctant to take at face value Socrates' comment at 39icio-di
that Hermogenes must learn about orthotes onomaton qua etymologizing "from
Homer and the other poets." Interpreters' dismissal of this observation regularly
involves their noting the hostility that Plato voices elsewhere toward the literary
tradition, notably in the Republic. I have argued that the consultation of literary
evidence is essential to the interpretation of what Plato has to say in the Cratylus,
that in fact this tradition serves as his most direct and central opponent there
with respect to etymologizing. In the most general way, etymology is after some-
thing promising, from Plato's perspective, insofar as it involves a strong concern
with questions of appropriateness and a firm belief that there are powerful con-
nections between elements of language and those of reality. It goes wrong, how-
ever, in the manner in which it expresses these commitments. According to Plato,
the literary tradition's practice of etymology goes astray by virtue of its adher-
ence to the view that the descriptive content of onomata, if rightly analyzed, it-

2. On Plato's uncompromising harshness toward art, see also Danto 1986, 6, 9, 194.
1/0 THE ANCIENT QUARREL

self discloses the phuseis of their referents and can therefore be used as a vital
source of insight into those natures. The dominant assumption of this activity,
that natures belong first and foremost to individuals, is also deeply problematic
on his accounting.
While the Cratylus concentrates largely on debunking etymology as a tool
by which one may gain genuine access to reality, it also points in the direction
of a more profitable orientation toward orthotes onomaton. As I have argued,
Plato's remarks at the close of the dialogue (438-440) pave the way for further,
positive reflections about appropriateness that have the theory of Forms as their
reference point and foundation. When Plato seeks constructions of orthotes
onomaton that are more fruitful from the vantage point of his construction of
reality, in the sense of meshing tightly therewith, devices other than etymology
take center stage. As we saw in chapter 4, Plato's own theories about orthotes
onomaton in the Phaedo and Republic 5 constitute revised versions of approaches
that figure prominently in the literary tradition. Seeking linguistic tools to re-
flect and reinforce his account of natures, in the Phaedo Plato embraces eponymy,
while in Republic 5 he concentrates on analyses of functional terms. The fact
that Plato turns to literary sources when seeking effective tools for the convey-
ance of his views on linguistic issues is unmistakable evidence of his willingness,
not simply to engage deeply with them in the role of critic, but also to embrace
those antecedents, where appropriate, as the origin of ideas that are themselves
worth endorsing.
If, as many have asserted, the Republic's famous discussion of poets culmi-
nates in their expulsion from the ideal polis, then the claims made here about
Plato's views concerning orthotes onomaton would not harmonize with his own
conclusion in that pivotal discussion. In fact, Plato's position in Republic 2-3
and 10 fits well with the attitude toward literary sources that is exhibited in his
discussions of orthotes onomaton. As I argued in chapter 5, in his delineation of
the ideal polis, multiple considerations, among them a recognition of the talent
of select individuals, prompt Plato to enlist poets as allies in the promotion of
human flourishing with respect both to early education and to the fostering of
unity on a range of civic occasions. Thus, in addition to including revised ver-
sions of devices prominent in literary sources among his own theories, Plato grants
that poetry per se has a role in that sound educational program without which
his community could not emerge and in reinforcing the solidarity of the polis
itself.
Although etymology does not, indeed cannot, become for Plato a tool of
philosophical theorizing, a philosophically motivated concern with the descrip-
tive content of onomata is evident in the Republic's discussion of early educa-
tion. As Plato indicates there, with a view toward fostering children's education
poets will invoke names' semantic constitution in ways that reinforce the posi-
tive depiction of desirable role models in their compositions. Moreover, as we
saw in chapter 2, the Cratylus' etymologies of divine names themselves reflect
the Republic's strictures about the characterization of divinities generally and
 CONCLUSION 1/1

names' descriptive content specifically. The two discussions thus sustain a comple-
mentary relationship: the Republic insists that names having a salutary impact
on children's attitudes be included in poetic compositions, and the Cratylus' ety-
mologies of divine onomata provide examples of what the right approach to these
names involves. Hence, the Cratylus—if viewed in light of the Republic—turns
out to supply material that may be used, if in a limited way, to address Plato's con-
cern with the negative impact of exposure to etymologies on individuals' well-being.
Some of those who have not concurred with what is alleged to be Plato's ex-
tremely harsh assessment of literature have concluded that the only way to de-
fend it from his condemnation is to diverge sharply from his approach to the
cognitive state, domain, and activity of its practitioners. Other supporters of poetry
have suggested that at least certain of the ingredients on the basis of which a
defense could be constructed (notably, mimesis and inspiration) are available —
though of course not developed as required —in Plato's writings.3 To the extent
that one accepts the account of Plato's own practice and theories offered in the
present study, one will be forced to reconsider the nature of his role in the Western
tradition with respect to the division between philosophy and literature.
Murdoch protests (1977, 77), against Plato, that "surely great art points in the
direction of the good and is at least more valuable to the moralist as an auxiliary
than dangerous as an enemy." I hope to have shown that one need not aban-
don—or even transform —Plato to get a defense of literature's potential value to
philosophy in salient arenas going.4 What must not be forgotten, above all, is
the open-minded stance that Plato displays toward sources of ideas. Plato's atti-
tude and methodology as characterized in this study in fact instantiate his own
message in the Phaedrus (27513-0), where Socrates chides his interlocutor, from
whom the dialogue takes its name, for paying heed to the identity of a speaker
rather than caring solely about the truth-content of what is said.5 While, on a
specific level, Plato's project as it involves literary sources differs from those of
contemporary philosophers, his general stance reflects more common ground
with present work than has been recognized. Underscoring Plato's attitude and
approach to his sources in the manner undertaken here serves as a reminder of
the fact that, far from serving merely as an adversary and target for those seeking
fruitful ways in which philosophy can benefit from appeals to literature, his work
represents an important precedent for inquiry in this area.

3. As a contemporary example of this type, one may cite Schaper 1968, 55: "Defending art
against the Platonic charges . . . means defending it with weapons which Plato thought of [in the
Ion], though did not himself use." On this general way of supporting poetry in antiquity and there-
after, see the discussions and bibliographical references in Murray 1996, 24-32; Halliwell i993b,
13-16, 25-27 nn. 20-31.
4. For recent emphasis on the importance of literature to reflection on, and discourse about,
human values and flourishing, see Nussbaum 1990, 1995; Williams 1993.
5. Cf. Woodruffs citing of this passage (1998, 524) in remarks on Plato's quotation of poets.
This page intentionally left blank
 BIBLIOGRAPHY

Adam,}., ed. 1902. Plato, Republic. Vols. 1-2. Cambridge: Cambridge University Press.
Allan, D.}. 1954. "The Problem of Cratylus." American Journal of Philology 75: 271-87.
Allen, J. 1994. "Failure and Expertise in the Ancient Conception of an Art." In Scien-
tific Failure, 81-108. Ed. T. Horowitz and A. I. Janis. Lanham, Md.: Rowman and
Littlefield.
Allen, R. E. 19653. "Participation and Predication in Plato's Middle Dialogues." In Allen
1965^ 43-60. Originally published in Philosophical Review 69 (1960): 147-64.
, ed. 1965^ Studies in Plato's Metaphysics. London: Routledge and Kegan Paul.
Allen, T. W., ed. 1917-1919. Homer, Opera. Vols. 3-4: Odyssey. 26 ed. Oxford: Clarendon
Press.
Anagnostopoulos, G. 1972. "Plato's Cratylus: The Two Theories of the Correctness of
Names." Review of Metaphysics 25: 691-736.
. 1973. "The Significance of Plato's Cratylus." Review of Metaphysics 27: 318-45.
Annas, }., tr. 1976. Aristotle, Metaphysics, Books M and N. Clarendon Aristotle Series.
Oxford: Clarendon Press.
. 1981. An Introduction to Plato's "Republic." Oxford: Clarendon Press.
. 1982. "Plato on the Triviality of Literature." In Plato on Beauty, Wisdom, and the
Arts, 1-28. Ed. J. Moravcsik and P. Temko. Totowa, N.J.: Rowman and Littlefield.
Arieti, J. A. 1991. Interpreting Plato: The Dialogues as Drama. Savage, Md.: Rowman
and Littlefield.
Arnim, H. von. 1912. Sprachliche Forschungen zur Chronologie der platonischen Dialoge.
Sitzungsberichte der Kaiserlichen Akademie der Wissenschaffen in Wien, Philoso-
phisch-historische Klasse, 169/3: 1-235. Vienna: Alfred Holder.
. 1929. Die Sprachliche Forschung als Grundlage der Chronologie der platonischen
Dialoge und der "Kratylos." Sitzungsberichte der Akademie der Wissenschaften
in Wien, Philosophisch-historische Klasse, 210/4: 1-26. Vienna: Holder-Pichler-
Tempsky.
Asmis, E. 1992. "Plato on Poetic Creativity." In The Cambridge Companion to Plato,
338-64. Ed. R. Kraut. Cambridge: Cambridge University Press.
Athanassakis, A. N., tr. 1976. The Homeric Hymns. Baltimore: Johns Hopkins Univer-
sity Press.

173
1/4 BIBLIOGRAPHY

Austin, N. 1972. "Name Magic in the Odyssey." California Studies in Classical Antiq-
uity 5: 1-19.
Bambrough, R. 1971. "Plato's Political Analogies." In Plato: A Collection of Critical
Essays, vol. 2: Ethics, Politics, and Philosophy of Art and Religion, 187-205. Ed.
G. Vlastos. Notre Dame: University of Notre Dame Press.
Barlow, S. A., tr. 1986. Euripides, Trojan Women. Warminster, England: Aris and
Phillips.
Barney, R. 1997. "Plato on Conventionalism." Phronesis 42: 143-62.
. 1998. "Socrates Agonistes: The Case of the Cratylus Etymologies." Oxford Stud-
ies in Ancient Philosophy 16: 63-98.
Barwick, K. 1957. Problems der stoischen Sprachlehre und Rhetorik. Abhandlungen der
sachsischen Akademie der Wissenschaften zu Leipzig, Philologisch-historische
Klasse, 49/3. Berlin: Akademie-Verlag.
Baxter, T. M. S. 1992. The "Cratylus": Plato's Critique of Naming. Leiden: E. J. Brill.
Belfiore, E. 1981. "The Role of the Visual Arts in Plato's Ideal State." Journal of the Theory
and Criticism of the Visual Arts i: 115-27.
. 1984. "A Theory of Imitation in Plato's Republic." Transactions of the American
Philological Association 114: 121-46.
Benardete, S. 1981. "Physics and Tragedy: On Plato's Cratylus." Ancient Philosophy
i: 127-40.
Benardete, S. G., and D. Grene, trs. 1991. The Complete Greek Tragedies. Ed. D. Grene
and R. Lattimore. Aeschylus, vol. 2: The Suppliant Maidens, Persians, Seven against
Thebes, Prometheus Bound. 2d ed. Chicago: University of Chicago Press.
Bestor, T. W. 1978. "Common Properties and Eponymy in Plato." Philosophical Quar-
terly 28: 189-207.
Blass, F., ed. 1898-1899. Isocrates, Orationes. Vols. 1-2. Leipzig: Teubner.
Bloom, A., tr. 1991. Plato, Republic. 2d ed. New York: Basic Books.
Bluck, R. S., tr. 1955. Plato, Phaedo. London: Routledge and Kegan Paul.
Bostock, D. 1986. Plato's "Phaedo." Oxford: Clarendon Press.
Bowra, C. M., ed. 1947. Pindar, Carmina cum Fragmentis. 26 ed. Oxford: Clarendon
Press.
Brandwood, L. 1990. The Chronology of Plato's Dialogues. Cambridge: Cambridge
University Press.
Broadhead, H. D., ed. 1960. Aeschylus, Persae. Cambridge: Cambridge University Press.
Brown, C. S. 1966. "Odysseus and Polyphemus: The Name and The Curse." Compara-
tive Literature 18: 193-202.
Burkert, W. 1968. "Orpheus und die Vorsokratiker: Bemerkungen zum Derveni-Papyrus
und zur pythagoreischen Zahlenlehre." Antike und Abendland 14: 93-114.
. 1970. "La genese des choses et des mots: Le papyrus de Derveni entre Anaxagore
et Cratyle." Etudes philosophiques 25: 443-55.
. 1983. "Eraclito nel papiro di Derveni: Due nuove testimonianze." In Atti del
Symposium Heracliteum 1981, vol. i, 37-42. Ed. L. Rossetti. Rome: Edizioni
dell'Ateneo.
. 1985. Greek Religion. Tr. J. Raffan. Cambridge, Mass.: Harvard University Press.
. 1986. "Der Autor von Derveni: Stesimbrotos Ilepi TeXeTrov?" Zeitschrift fur
Papyrologie und Epigraphik 62: 1-5.
. 1997. "Star Wars or One Stable World? A Problem of Presocratic Cosmogony
(PDerv. Col. XXV)." In Laks and Most 1997, 167-74.
 BIBLIOGRAPHY 175

Burnet,}., ed. 1901-1907. Plato, Opera. Vols. 2-5. Oxford: Clarendon Press.
, ed. 1911. Plato, Phaedo. Oxford: Clarendon Press.
Bywater, I., ed. 1894. Aristotle, Ethica Nicomachea. Oxford: Clarendon Press.
Caizzi, F. D., ed. 1966. Antisthenis Fragments. Milan: Istituto Editoriale Cisalpino.
Calvert, B. 1970. "Forms and Flux in Plato's Cratylus." Phronesis 15: 26-47.
Capelle, W. 1922. "Zur hippokratischen Frage." Hermes 57: 247-65.
Cherniss, H. 1962. Aristotle's Criticism of Plato and the Academy. New York: Russell
and Russell.
Code, A. 1986. "Aristotle: Essence and Accident." In Philosophical Grounds of Ratio-
nality: Intentions, Categories, Ends, 411-39. Ed. R. E. Grandy and R. Warner.
Oxford: Clarendon Press.
Collingwood, R. G. 1925. "Plato's Philosophy of Art." Mind, n.s., 34: 154-72.
. 1958. The Principles of Art. New York: Oxford University Press.
Cooke, E. F. 1999. "The Moral and Intellectual Development of the Philosopher in
Plato's Republic." Ancient Philosophy 19: 37-44.
Cooper, J. M. 1977. "The Psychology of Justice in Plato." American Philosophical Quar-
terly 14: 151-57.
Cornford, F. M. 1935. "A New Fragment of Parmenides." Classical Review 49: 122-
2
3-
, tr. 1941. Plato, Republic. London: Oxford University Press.
Craik, E. M. 1993. "AIAtlS in Euripides' Hippolytos 373-430: Review and Reinterpre-
tation." Journal of Hellenic Studies 113: 45-59.
Cross, R. C., and A. D. Woozley. 1964. Plato's "Republic": A Philosophical Commen-
tary. London: Macmillan.
Curd, P. 1998. The Legacy of Parmenides: Eleatic Monism and Later Presocratic Thought.
Princeton: Princeton University Press.
Dahlmann, H. 1929. Review of Warburg 1929. Deutsche Literaturzeitung fur Kritik der
internationalen Wissenschaft, n.s., 6 (August 31): 1671-76.
Dalimier, C., tr. 1998. Plato, Cratyle. Paris: Flammarion.
Danto, A. C. 1986. The Philosophical Disenfranchisement of Art. New York: Columbia
University Press.
Demand, N. 1975. "The Nomothetes of the Cratylus." Phronesis 20: 106-9.
Derbolav,}. 1972. Platans Sprachphilosophie im "Kratylos" und in den spdteren Schriften.
Darmstadt: Wissenschaftliche Buchgesellschaft.
Diels, H. 1910. "Die Anfange der Philologie bei den Griechen." Neue Jahrbucher fur
das klassische Altertum 25: 1-25.
. [1951-1952] 1992. Die Fragmente der Vorsokratiker. Ed. W. Kranz. Vols. 1-2. 6th
ed. Reprint, Zurich: Weidmann.
Diggle, J., ed. [1981] 1986. Euripides, Fabulae. Vol. 2. Reprinted with corrections, Ox-
ford: Clarendon Press.
, ed. [1984] 1989. Euripides, Fabulae. Vol. i. Reprinted with corrections, Oxford:
Clarendon Press.
Dimock, G. E., Jr. 1956. "The Name of Odysseus." Hudson Review 9: 52-70.
Dittmeyer, L., ed. 1907. Aristotle, De Animalibus Historia. Leipzig: Teubner.
Dodds, E. R., ed. 1959. Plato, Gorgias. Oxford: Clarendon Press.
Dover, K. J., ed. 1970. Aristophanes, Clouds. Abridged ed. Oxford: Oxford University
Press.
, ed. 1993. Aristophanes, Frogs. Oxford: Clarendon Press.
1/6 BIBLIOGRAPHY

Duke, E. A., W. F. Hicken, W. S. M. Nicoll, D. B. Robinson, and J. C. G. Strachan,

eds. 1995. Plato, Opera. Vol. i. Oxford: Clarendon Press.
Edmundson, M. 1995. Literature against Philosophy, Plato to Derrida: A Defence of
Poetry. Cambridge: Cambridge University Press.
Edwards, M. J. 1991. "Notes on the Derveni Commentator." Zeitschrift fur Papyrologie
und Epigraphik 86: 203-11.
Elias, J. A. 1984. Plato's Defence of Poetry. Albany: State University of New York Press.
Else, G. F. 1958. "'Imitation' in the Fifth Century." Classical Philology 53: 73-90.
. 1972. The Structure and Date of Book 10 of Plato's "Republic." Heidelberg: Carl
Winter Universitatsverlag.
Evelyn-White, H. G., tr. 1914. Hesiod, Homeric Hymns and Homerica. Loeb Classical
Library. Cambridge, Mass.: Harvard University Press.
Faraone, C. A. 1985. "Aeschylus' v^ivoc, 6eo|j,io<; (Eum. 306) and Attic Judicial Curse
Tablets." Journal of Hellenic Studies 105: 150-54.
9. "An Accusation of Magic in Classical Athens (Ar. Wasps 946-48)." Trans-
actions of the American Philological Association 119: 149-60.
. 1991. "The Agonistic Context of Early Greek Binding Spells." In Magika Hiera:
Ancient Greek Magic and Religion, 3-32. Ed. C. A. Faraone and D. Obbink. New
York: Oxford University Press.
Ferguson, J. 1963. "Sun, Line, and Cave Again." Classical Quarterly, n.s., 13: 188-93.
Ferrari, G. R. F. 1989. "Plato and Poetry." In The Cambridge History of Literary Criti-
cism, vol. i: Classical Criticism, 92-148. Ed. G. A. Kennedy. Cambridge: Cambridge
University Press.
Ferwerda, R. 1985. "The Meaning of the Word orojia in Plato's Cratylus 4OoC." Hermes
113: 266-79.
Fine, G. 1978. "Knowledge and Belief in Republic V." Archiv fur Geschichte der
Philosophie 60: 121-39.
. 1993. On Ideas: Aristotle's Criticism of Plato's Theory of Forms. Oxford: Clarendon
Press.
Fitzgerald, R., tr. 1941. Sophocles, Oedipus at Colonus. New York: Harcourt Brace.
, tr. 1963. Homer, Odyssey. Garden City, N.Y.: Anchor Books/Doubleday.
Fogelin, R. J. 1971. "Three Platonic Analogies." Philosophical Review 80: 371-82.
Fortenbaugh, W. W. 1975. "Plato: Temperament and Eugenic Policy." Arethusa 8: 283-
305.
Fowler, H. N., tr. 1926. Plato. Vol. 4. Loeb Classical Library. Cambridge, Mass.: Harvard
University Press.
Fraenkel, E., ed. 1950. Aeschylus, Agamemnon. Vol. 2. Oxford: Clarendon Press.
Frege, G. 1950. The Foundations of Arithmetic. Tr. J. L. Austin. Oxford: Blackwell.
Friedlander, P. 1964. Plato. Vol. 2. Tr. H. Meyerhoff. New York: Bollingen Founda-
tion.
. 1969. Plato. Vol. 3. Tr. H. Meyerhoff. Princeton: Princeton University Press.
Funghi, M. S. 1997. "The Derveni Papyrus." In Laks and Most 1997, 25-37.
Gaiser, K. 1974. Name und Sache in Platans "Kratylos." Heidelberg: Carl Winter
Universitatsverlag.
Gallop, D., tr. [1975] 1988. Plato, Phaedo. Clarendon Plato Series. Reprinted with cor-
rections, Oxford: Clarendon Press.
Golden, L. 1975. "Plato's Concept of Mimesis." British Journal of Aesthetics 15: 118-31.
Goldschmidt, V. 1940. Essai sur le "Cratyle": Contribution a I'histoire de la pensee de Platan.
 BIBLIOGRAPHY 177

Bibliotheque de 1'Ecole des Hautes Etudes, sciences historiques et philologiques,

fascicule 279. Paris: Librairie Ancienne Honore Champion.
Gonzalez, F. ]., ed. 1995. The Third Way: New Directions in Platonic Studies. Lanham,
Md.: Rowman and Littlefield.
8a. Dialectic and Dialogue: Plato's Practice of Philosophical Inquiry. Evanston,
111.: Northwestern University Press.
. 19980. "Nonpropositional Knowledge in Plato." Apeiron 31: 235-84.
Gordon, J. 1999. Turning toward Philosophy: Literary Device and Dramatic Structure
in Plato's Dialogues. University Park: Pennsylvania State University Press.
Gottschalk, H. B. 1980. Heraclides ofPontus. Oxford: Clarendon Press.
Gould, J. 1955. The Development of Plato's Ethics. Cambridge: Cambridge University Press.
Gould, T. 1990. The Ancient Quarrel between Poetry and Philosophy. Princeton: Princeton
University Press.
Graham, D. W. 1991. "Socrates, the Craft Analogy, and Science." Apeiron 24: 1-24.
Greene, W. C. 1918. "Plato's View of Poetry." Harvard Studies in Classical Philology
29: 1-75.
Grene, D., tr. 1991. The Complete Greek Tragedies. Ed. D. Grene and R. Lattimore.
Sophocles, vol. i: Oedipus the King, Oedipus at Co/onus, Antigone. 2d ed. Chi-
cago: University of Chicago Press.
Grey, D. R. 1952. "Art in the Republic." Philosophy 27: 291-310.
Griffith, M. 1978. "Euripides Alkestis 636-641." Harvard Studies in Classical Philology
82: 83-86.
, ed. 1983. Aeschylus, Prometheus Bound. Cambridge: Cambridge University Press.
Griswold, C. 1981. "The Ideas and the Criticism of Poetry in Plato's Republic, Book
10." Journal of the History of Philosophy 19: 135-50.
, ed. 1988. Platonic Writings, Platonic Readings. New York: Routledge.
Grote, G. 1867. Plato and the Other Companions of Sokrates. Vol. 2. 2cl ed. London:
John Murray.
Grube, G. M. A. 1927. "The Marriage Laws in Plato's Republic." Classical Quarterly 21:
95-99-
. [1968] 1995. The Greek and Roman Critics. Reprint, Indianapolis: Hackett.
, tr. 1974. Plato, Republic. Indianapolis: Hackett.
Guthrie, W. K. C. 1965. A History of Greek Philosophy. Vol. 2: The Presocratic Tradi-
tion from Parmenides to Democritus. Cambridge: Cambridge University Press.
. 1971. The Sophists. Cambridge: Cambridge University Press.
. 1975. A History of Greek Philosophy. Vol. 4: Plato, the Man and His Dialogues:
Earlier Period. Cambridge: Cambridge University Press.
. 1978. A History of Greek Philosophy. Vol. 5: The Later Plato and the Academy.
Cambridge: Cambridge University Press.
Hackforth, R., tr. 1955. Plato, Phaedo. Cambridge: Cambridge University Press.
Hall, F. W., and W. M. Geldart, eds. 1906-1907. Aristophanes, Comoediae. Vols. 1-2.
2d ed. Oxford: Clarendon Press.
Hall, R. W. 1971. "Techne and Morality in the Gorgias." In Essays in Ancient Greek
Philosophy, 202-18. Ed. J. P. Anton and G. L. Kustas. Albany: State University of
New York Press.
. 1974. "Plato's Theory of Art: A Reassessment." American Society for Aesthetics
33: 75-82.
Halliwell, S., tr. 19933. Plato, Republic 5. Warminster, England: Aris and Phillips.
1/8 BIBLIOGRAPHY

, tr. iggjb. Plato, Republic 10. 2d corr. impr. Warminster, England: Aris and
Phillips.
Hartland-Swann, J. 1951. "Plato as Poet: A Critical Interpretation." Philosophy 26: 3-18.
Havelock, E. A. 1963. Preface to Plato. Cambridge, Mass.: Harvard University Press.
. 1986. The Muse Learns to Write: Reflections on Orality and Literacy from Antiq-
uity to the Present. New Haven: Yale University Press.
Heath, D. D. 1888. "On Plato's Cratylus." Journal of Philology 17: 192-218.
Heinimann, F. 1976. "Erne vorplatonische Theorie der te,yyr\." In Wege der Forschung,
vol. 187: Sophistik, 127-69. Ed. C. J. Classen. Darmstadt: Wissenschaftliche
Buchgesellschaft.
Henry, M. 1986. "The Derveni Commentator as Literary Critic." Transactions of the
American Philological Association 116: 149-64.
Henry, P., and H.-R. Schwyzer, eds. 1964-1977. Plotinus, Opera. Vols. 1-2. Oxford:
Clarendon Press.
Heubeck, A., S. West, and J. B. Hainsworth. 1988. A Commentary on Homer's "Odys-
sey." Vol. i: Introduction and Books I-VIII. Oxford: Clarendon Press.
Hicks, R. D., tr. 1931. Diogenes Laertius, Lives of Eminent Philosophers. Vol. 2. Loeb
Classical Library. Cambridge, Mass.: Harvard University Press.
Higbie, C. 1995. Heroes' Names, Homeric Identities. Albert Bates Lord Studies in Oral
Tradition, vol. 10. New York: Garland.
Hirschle, M. 1979. Sprachphilosophie und Namenmagie im Neuplatonismus: Mit einem
Exkurs zu "Demokrit" B 142. Beitra'ge zur klassischen Philologie, ed. E. Heitsch,
R. Merkelbach, and C. Zintzen, vol. 96. Meisenheim: Verlag Anton Hain.
Hogan, R. 1977. "The Techne Analogy in the Charmides." Philosophy Research Archives
3: 702-8.
Hude, C., ed. 1927. Herodotus, Historiae. Vols. 1-2. 3d ed. Oxford: Clarendon Press.
Irwin, T. H. 19773. "Plato's Heracleiteanism." Philosophical Quarterly 27: 1-13.
. i977b. Plato's MoralTheory: The Early and Middle Dialogues. Oxford: Clarendon
Press.
-. 1995. Plato's Ethics. New York: Oxford University Press.
Jaeger, W. 1945. Paideia: The Ideals of Greek Culture. Vol. r. Archaic Greece: The Mind
of Athens. Tr. G. Highet. 2d ed. New York: Oxford University Press.
, ed. 1957. Aristotle, Metaphysica. Oxford: Clarendon Press.
Janaway, C. 1992. "Craft and Fineness in Plato's Ion." Oxford Studies in Ancient Phi-
losophy 10: 1-23.
. 1995. Images of Excellence: Plato's Critique of the Arts. Oxford: Clarendon Press.
Jevons, F. B. 1908. "Graeco-Italian Magic." In Anthropology and the Classics: Six Lec-
tures Delivered before the University of Oxford, 93-120. Ed. R. R. Marett. Oxford:
Clarendon Press.
Jordan, D. 1980. "Two Inscribed Lead Tablets from a Well in the Athenian Kerameikos."
Mitteilungen des deutschen archdologischen Institute, Athenische Abteilung 95: 225-
39-
. 1985. "A Survey of Greek Defixiones Not Included in the Special Corpora."
Greek, Roman, and Byzantine Studies 26: 151-97.
. 1988. "New Archaeological Evidence for the Practice of Magic in Classical Ath-
ens." In Praktika of the 12th International Congress of Classical Archaeology, Ath-
ens, September 1983, vol. 4, 273-77. Athens: Tameio Archaiologikon Poron kai
Apallotrioseon.
 BIBLIOGRAPHY 1/9

Jowett, B., tr. 1892. The Dialogues of Plato. Vols. 1-2. 3d ed. London: Oxford University
Press.
Kahn, C. H. 1973. "Language and Ontology in the Cratylus." In Exegesis and Argu-
ment: Studies in Greek Philosophy Presented to Gregory Vlastos, 1.52-76. Ed. E. N.
Lee, A. P. D. Mourelatos, and R. M. Rorty. New York: Humanities Press.
. 1979. The Art and Thought ofHeraclitus: An Edition of the Fragments with Trans-
lation and Commentary. Cambridge: Cambridge University Press.
. 1985. "The Beautiful and the Genuine: A Discussion of Paul Woodruff, Plato,
Hippias Major." Oxford Studies in Ancient Philosophy 3: 261-87.
. 1986. "Les mots et les formes dans le Cratyle de Platon." Cahiers de philosophie
ancienne 5: 91-103.
. 1993. "Plato's Ion and the Problem olTechne." In Nomodeiktes: Greek Studies in
Honor of Martin Ostwald, 369-78. Ed. R. M. Rosen and J. Farrell. Ann Arbor:
University of Michigan Press.
. 1996. Plato and the Socratic Dialogue: The Philosophical Use of a Literary Form.
Cambridge: Cambridge University Press.
. 1997. "Was Euthyphro the Author of the Derveni Papyrus?" In Laks and Most
J
997. 55-63-
Kamtekar, R. 1998. "Imperfect Virtue." Ancient Philosophy 18: 315-39.
Kassel, R., ed. [1965] 1966. Aristotle, De Arte Poetica Liber. Reprinted with corrections,
Oxford: Clarendon Press.
Kato, M. 1986. Techne und Philosophie bei Platon. Frankfurt am Main: Peter Lang.
Kerferd, G. B. 1981. The Sophistic Movement. Cambridge: Cambridge University Press.
Kirk, G. S. 1951. "The Problem of Cratylus." American Journal of Philology 72: 225-53.
. 1962. Heraclitus: The Cosmic Fragments. Cambridge: Cambridge University
Press.
. 1985. The "Iliad": A Commentary. Vol. i: Books 1-4. Cambridge: Cambridge
University Press.
. 1989. "Homer." In The Cambridge History of Classical Literature, vol. i, pt. i:
Early Greek Poetry, 1-50. Ed. P. E. Easterling and B. M. W. Knox. Cambridge:
Cambridge University Press.
Kirk, G. S., J. E. Raven, and M. Schofield. 1983. The Presocratic Philosophers: A Criti-
cal History with a Selection of Texts. 2d ed. Cambridge: Cambridge University Press.
Kranz, W. 1933. Stasimon: Untersuchungen zu Form und Gehalt der griechischen
Tragodie. Berlin: Weidmannsche Buchhandlung.
Kraut, R. 1973. "Egoism, Love, and Political Office in Plato." Philosophical Review 82:
330-44.
. 1987. Review of Elias 1984. Nous 21: 66-69.
. 1992. "The Defense of Justice in Plato's Republic." In The Cambridge Compan-
ion to Plato, 311-37. Ed. R. Kraut. Cambridge: Cambridge University Press.
. 1993. "Return to the Cave: Republic 519-521." Proceedings of the Boston Area
Colloquium in Ancient Philosophy j: 43-62.
Kretzmann, N. 1967. "Semantics, History of." In The Encyclopedia of Philosophy, vol. 7,
358-406. New York: Macmillan.
. 1971. "Plato on the Correctness of Names." American Philosophical Quarterly 8:
126-38.
Kube, J. 1969. Te^vri und 'Apetri: Sophistisches und platonisches Tugendwissen. Berlin:
De Gruyter.
180 BIBLIOGRAPHY

Laks, A. 1997. "Between Religion and Philosophy: The Function of Allegory in the
Derveni Papyrus." Phronesis 42: 121-42.
Laks, A., and G. W. Most. 1997. Studies on the Derveni Papyrus. Oxford: Clarendon
Press.
Lattimore, R., tr. 1951. Homer, Iliad. Chicago: University of Chicago Press.
, tr. 1969. Aristophanes, Frogs. In Aristophanes, Four Comedies. Ed. W. Arrowsmith.
Ann Arbor Paperbacks. Ann Arbor: University of Michigan Press.
Lattimore, R., S. G. Benardete, and D. Grene, trs. 1953-1956. The Complete Greek Trag-
edies. Ed. D. Grene and R. Lattimore. Aeschylus, vols. 1-2. Chicago: University of
Chicago Press.
Lattimore, R., R. Warner, R. Gladstone, D. Grene, W. Arrowsmith, W. Bynner, J. F.
Nims, R. F. Willetts, F. W. Jones, C. R. Walker, E. T. Vermeule, and E. Wyckoff,
trs. 1955-1959. The Complete Greek Tragedies. Ed. D. Grene and R. Lattimore.
Euripides, vols. 1-5. Chicago: University of Chicago Press.
Leky, M. [1919] 1967. Plato als Sprachphilosoph: Wiirdigung des platonischen "Kratylus."
Reprint, New York: Johnson Reprint Corporation.
Lesses, G. 1986. "Plato's Lysis and Irwin's Socrates." International Studies in Philoso-
phy 18: 33-43.
Levin, S. B. 1995. "What's in a Name? A Reconsideration of the Cratylus' Historical
Sources and Topics." Ancient Philosophy 15: 91-115.
. 19963. "Platonic Eponymy and the Literary Tradition." Phoenix 50: 197-207.
. i9g6b. "Women's Nature and Role in the Ideal Palis: Republic 5 Revisited." In
Feminism and Ancient Philosophy, 13-30. Ed. J. K. Ward. New York: Routledge.
. 1997. "Greek Conceptions of Naming: Three Forms of Appropriateness in Plato
and the Literary Tradition." Classical Philology 92: 46-57.
. 2000. Commentary on Catherine Osborne, "'No' Means Yes': The Seduction
of the Word in Plato's Phaedrus." Proceedings of the Boston Area Colloquium in
Ancient Philosophy 15: 282-92.
Levinson, R. B. 1957. "Language and the Cratylus: Four Questions." Review of Meta-
physics n: 28-41.
Liddell, H. G., and R. Scott. 1940. A Greek-English Lexicon, gth ed. Revised and aug-
mented by H. S. Jones. With a revised supplement, 1996. Oxford: Clarendon Press.
Lloyd, G. E. R. 1987. The Revolutions of Wisdom: Studies in the Claims and Practice of
Ancient Greek Science. Sather Classical Lectures, vol. 52. Berkeley: University of
California Press.
Lloyd-Jones, H., and N. G. Wilson, eds. 1990. Sophocles, Fabulae. Oxford: Clarendon
Press.
Lodge, R. C. 1953. Plato's Theory of Art. London: Routledge and Kegan Paul.
Lombardo, S., tr. 1993. Hesiod, Worfes and Days and Theogony. Introduction, Notes,
and Glossary by R. Lamberton. Indianapolis: Hackett.
Looy, H. van. 1975. "Figura etymologica et etymologic dans 1'ceuvre de Sophocle."
Museum Philologum Londiniense i: 109-19.
Lorenz, K., and J. Mittelstrass. 1967. "On Rational Philosophy of Language: The
Programme in Plato's Cratylus Reconsidered." Mind, n.s., 76: 1-20.
Lorimer, W. L., ed. 1933. Aristotle, De Mundo. Paris: Societe d'edition "Les belles
lettres."
Luce, J. V. 1964. "The Date of the Cratylus." American Journal of Philology 85: 136-54.
. 1965. "The Theory of Ideas in the Cratylus." Phronesis 10: 21-36.
 BIBLIOGRAPHY l8l

. 1969. "Plato on Truth and Falsity in Names." Classical Quarterly, n.s., 19: 222-32.
Lulofs, H. J. D., ed. 1965. Aristotle, De Generations Animalium. Oxford: Clarendon Press.
MacDowell, D. M, ed. and tr. 1982. Gorgias, Encomium of Helen. Bristol: Bristol Clas-
sical Press.
Mackenzie, M. M. 1986. "Putting the Cratylus in Its Place." Classical Quarterly, n.s.,
36: 124-50.
Malcolm, J. 1981. "Semantics and Self-Predication in Plato." Phronesis 26: 286-94.
. 1985. "Vlastos on Pauline Predication." Phronesis 30: 79-91.
. 1991. Plato on the Self-Predication of Forms: Early and Middle Dialogues. Ox-
ford: Oxford University Press.
Marchant, E. C., ed. 1921. Xenophon, Opera Omnia. Vol. 2. 2d ed. Oxford: Clarendon
Press.
Marcovich, M. 1967. Heraclitus: Greek Text with a Short Commentary. Merida, Vene-
zuela: Los Andes University Press.
Marrou, H.-I. 1981. Histoire de {'education dans I'Antiquite. Vol. i: Ee monde grec. 7th
ed. Paris: Editions du Seuil.
McCabe, M. M. 1992. "Myth, Allegory and Argument in Plato." In The Language of
the Cave, 47-67. Ed. A. Barker and M. Warner. Apeiron 25, no. 4. Edmonton, Alb.:
Academic Printing and Publishing.
McKeon, R. 1936. "Literary Criticism and the Concept of Imitation in Antiquity."
Modern Philology 34: 1-35.
McKirahan, R. D., Jr. 1994. Philosophy before Socrates: An Introduction with Texts and
Commentary. Indianapolis: Hackett.
Meinwald, C. C. 1991. Plato's "Parmenides." New York: Oxford University Press.
Meridier, L., ed. and tr. 1931. Platon: GLuvres completes. Vol. 5. Paris: Societe d'edition
"Les belles lettres."
Merkelbach, R. 1967. "Der orphische Papyrus von Derveni." Zeitschrift fur Papyrologie
und Epigraphik i: 21-32.
Monro, D. B., and T. W. Allen, eds. 1920. Homer, Opera. Vols. 1-2: Iliad. 3d ed. Ox-
ford: Clarendon Press.
Moore, G. E., G. F. Stout, and G. D. Hicks. 1923. "Are the Characteristics of Particular
Things Universal or Particular?" Proceedings of the Aristotelian Society, supp.
vol. 3: 95-128.
Moore, ]., M. Jameson, and D. Grene, trs. 1957. The Complete Greek Tragedies. Ed. D.
Grene and R. Lattimore. Sophocles, vol. 2: Ajax, The Women ofTrachis, Electra,
Philoctetes. Chicago: University of Chicago Press.
Moraux, P, ed. and tr. 1965. Aristotle, Du del. Paris: Societe d'edition "Les belles lettres."
Moravcsik, J. 1963. "The 'Third Man' Argument and Plato's Theory of Forms." Phronesis
8: 50-62.
. 1976. "Recollecting the Theory of Forms." In Facets of Plato's Philosophy, 1-20.
Ed. W. H. Werkmeister. Phronesis, supp. vol. 2. Assen: Van Gorcum.
. 1991. "Appearance and Reality in Heraclitus' Philosophy." Monist 74: 551-66.
. 1992. Plato and Platonism: Plato's Conception of Appearance and Reality in
Ontology, Epistemology, and Ethics, and Its Modem Echoes. Oxford: Blackwell.
Morford, M. P. O., and R. J. Lenardon. 1995. Classical Mythology. 5th ed. White Plains,
N.Y.: Longman.
Murdoch, I. 1977. The Fire and the Sun: Why Plato Banished the Artists. Oxford:
Clarendon Press.
182 BIBLIOGRAPHY

Murphy, N. R. 1967. The Interpretation of Plato's "Republic." Oxford: Clarendon Press.

Murray, A. T., tr. 1919. Homer, Odyssey. Vols. 1-2. Loeb Classical Library. Cambridge,
Mass.: Harvard University Press.
Murray, G., ed. 1913. Euripides, Fahulae. Vol. 3. 2d ed. Oxford: Clarendon Press.
, tr. 1916. Aristophanes, Frogs. London: Longmans Green.
Murray, P. 1981. "Poetic Inspiration in Early Greece." Journal of Hellenic Studies 101:
87-100.
. 1992. "Inspiration and Mimesis in Plato." In The Language of the Cave, 27-46.
Ed. A. Barker and M. Warner. Apeiron 25, no. 4. Edmonton, Alb.: Academic Print-
ing and Publishing.
-, ed. 1996. Plato on Poetry: "Ion"; "Republic" 3766-398^19; "Republic" 595-608^10.
Cambridge Greek and Latin Classics. Cambridge: Cambridge University Press.
Nagy, G. 1974. Comparative Studies in Greek and Indie Meter. Cambridge, Mass.:
Harvard University Press.
. 1979. The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry.
Baltimore: Johns Hopkins University Press.
Nauck, A., ed. 1889. Tragicorum Graecorum Fragmenta. ad ed. Leipzig: Teubner.
Nehamas, A. 1973. "Predication and Forms of Opposites in the Phaedo." Review of
Metaphysics 26: 461-91.
. 1979. "Self-Predication and Plato's Theory of Forms." American Philosophical
Quarterly 16: 93-103.
. 1982. "Plato on Imitation and Poetry in Republic 10." In Plato on Beauty, Wis-
dom, and the Arts, 47-78. Ed. J. Moravcsik and P. Temko. Totowa, N.J.: Rowman
and Littlefield.
. 1988. "Plato and the Mass Media." Monist jr. 214-34.
Nehamas, A., and P. Woodruff, trs. 1989. Plato, Symposium. Indianapolis: Hackett.
Nettleship, R. L. 1937. Lectures on the "Republic" of Plato. iA ed. London: Macmillan.
Neustadt, E. 1929. "Wort und Geschehen in Aischylos' Agamemnon." Hermes 64: 243-
65.
Nightingale, A. W. 1995. Genres in Dialogue: Plato and the Construct of Philosophy.
Cambridge: Cambridge University Press.
Nisetich, F. J., tr. 1980. Pindar's Victory Songs. Baltimore: Johns Hopkins University
Press.
Nussbaum, M. C. 1972. ""FYXH in Heraclitus, I." Phronesis 17: 1-16.
. 1986. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philoso-
phy. Cambridge: Cambridge University Press.
. 1990. Love's Knowledge: Essays on Philosophy and Literature. New York: Oxford
University Press.
. 1995. Poetic Justice: The Literary Imagination and Public Life. Boston: Beacon
Press.
O'Brien, M. J. 1967. The Socratic Paradoxes and the Greek Mind. Chapel Hill: Univer-
sity of North Carolina Press.
Owen, G. E. L. 19863. "A Proof in the Peri Ideon." In Logic, Science, and Dialectic:
Collected Papers in Greek Philosophy, 165-79. Ed. M. Nussbaum. Ithaca: Cornell
University Press. Originally published in Journal of Hellenic Studies 77 (1957):
103-11.
. i986b. "Logic and Metaphysics in Some Earlier Works of Aristotle." In Logic,
Science, and Dialectic: Collected Papers in Greek Philosophy, 180—99. Ed. M.
 BIBLIOGRAPHY 183

Nussbaum. Ithaca: Cornell University Press. Originally published in Aristotle and

Plato in the Mid-Fourth Century, ed. I. During and G. E. L. Owen, Studia Graeca
et Latina Gothoburgensia 11 (Goteborg, 1960), 163-90.
Page, D., ed. 1972. Aeschylus, Tragoediae. Oxford: Clarendon Press.
Parke, H. W. 1977. Festivals of the Athenians. Ithaca: Cornell University Press.
Parry, R. D. 1983. "The Craft of Justice." In New Essays on Plato, 19-38. Ed. F. J. Pelletier
and }. King-Farlow. Guelph, Ont.: Canadian Association for Publishing in Phi-
losophy.
. 1996. Plato's Craft of Justice. Albany: State University of New York Press.
Penner, T. 1992. "Socrates and the Early Dialogues." In The Cambridge Companion to
Plato, 121-69. Ed. R. Kraut. Cambridge: Cambridge University Press.
Peradotto, J. 1990. Man in the Middle Voice: Name and Narration in the "Odyssey."
Princeton: Princeton University Press.
Pfeiffer, R. 1968. History of Classical Scholarship: From the Beginnings to the End of
the Hellenistic Age. Oxford: Clarendon Press.
Pickard-Cambridge, A. 1988. The Dramatic Festivals of Athens, zd ed. Revised by J. Gould
and D. M. Lewis. Oxford: Clarendon Press.
Press, G. A., ed. 1993. Plato's Dialogues: New Studies and Interpretations. Lanham, Md.:
Rowman and Littlefield.
Quincey, J. H. 1963. "Etymologica." Rheinisches Museum fur Philologie, n.s., 106: 142-48.
Rankin, H. D. 1965. "Plato's Eugenic EYO>HMIA and AnOOESE in Republic,
Book V." Hermes 93: 407-20.
Reeve, C. D. C. 1988. Philosopher-Kings: The Argument of Plato's "Republic." Princeton:
Princeton University Press.
, tr. 1998. Plato, Cratylus. Indianapolis: Hackett.
Rich, A. N. M. 1960. "Plotinus and the Theory of Artistic Imitation." Mnemosyne 13
(ser. 4): 233-39.
Richardson, N. J., ed. 1974. The "Homeric Hymn to Demeter." Oxford: Clarendon Press.
. 1993. The "Iliad": A Commentary. Vol. 6: Books 22-24. Cambridge: Cambridge
University Press.
Rijlaarsdam, J. C. 1978. Platon tiberdie Sprache: Ein Kommentarzum Kratylos, rnit einem
Anhang iiber die Quelle der Zeichentheorie Ferdinand de Saussures. Utrecht: Bohn,
Scheltema, and Holkema.
Robertson, D. S. 1924. "Aeschylea." Classical Review 38: 109-10.
Robin, L., ed. and tr. 1926. Platon: (Euvres completes. Vol. 4. Paris: Societe d'edition
"Les belles lettres."
Robinson, R. 1953. Plato's Earlier Dialectic. 2cl ed. Oxford: Clarendon Press.
. 1969. "The Theory of Names in Plato's Cratylus." In Essays in Greek Philoso-
phy, 100-117. Oxford: Clarendon Press. Originally published in Revue Internationale
de philosophic 9 (1955): 221-36.
Robinson, T. M. 1987. Heraclitus, Fragments: A Text and Translation with a Commen-
tary. Phoenix supp. vol. 22. Toronto: University of Toronto Press.
Roochnik, D. L. 1986. "Socrates's Use of the Techne-Analogy." Journal of the History
of Philosophy 24: 295-310.
. 1990. The Tragedy of Reason: Toward a Platonic Conception of Logos. New York:
Routledge.
. 1996. Of Art and Wisdom: Plato's Understanding ofTechne. University Park:
Pennsylvania State University Press.
184 BIBLIOGRAPHY

Rosenstock, B. 1992. "Father and Sons: Irony in the Cratylus." Arethusa 25: 385-417.
Ross, D. 1953. Plato's Theory of Ideas. Oxford: Clarendon Press.
. 1955. "The Date of Plato's Cratylus." Revue Internationale de philosophic
9: 187-96.
, ed. 1956. Aristotle, De Anima. Oxford: Clarendon Press.
, ed. [1958] 1989. Aristotle, Topica et Sophistici Elenchi. Reprinted with correc-
tions, Oxford: Clarendon Press.
ed. 1959. Aristotle, Ars Rhetorica. Oxford: Clarendon Press.
Rowe, C. 1997. "Plato: Aesthetics and Psychology." In Routledge History of Philoso-
phy, vol. i: From the Beginning to Plato, 425-55. Ed. C. C. W. Taylor. London:
Routledge.
Ruijgh, C. J. 1976. "Observations sur 1'emploi onomastique de KeKXfloGai vis-a-vis de
celui de Kakeics&m, notamment dans la tragedie attique." In Miscellanea tragica
in honorem }. C. Kamerbeek, 333-95. Ed.}. M. Bremer, S. L. Radt, and C. ]. Ruijgh.
Amsterdam: Adolf Hakkert.
Rumsey, W. D. 1987. "Plato in the Cratylus on Speaking, Language, and Learning."
History of Philosophy Quarterly 4: 385-403.
Russell, B. 1971. "On the Relations of Universals and Particulars." In Logic and Knowledge:
Essays, 1901-1950, 105-24. Ed. R. C. Marsh. Reprint, New York: Capricorn Books.
B. The Problems of Philosophy. Buffalo, N.Y.: Prometheus Books.
Rusten, }. S. 1985. "Interim Notes on the Papyrus from Derveni." Harvard Studies in
Classical Philology 89: 121-40.
Rutherford, R. B. 1995. The Art of Plato: Ten Essays in Platonic Interpretation. Cam-
bridge, Mass.: Harvard University Press.
Ryle, G. 1966. Plato's Progress. Cambridge: Cambridge University Press.
Sambursky, S. 1959. "A Democritean Metaphor in Plato's Kratylos." Phronesis 4: 1—4.
Sansone, D. 1996. "Plato and Euripides." Illinois Classical Studies 21: 35-67.
Sayre, K. M. 1995. Plato's Literary Garden: How to Read a Platonic Dialogue. Notre
Dame: University of Notre Dame Press.
Schaarschmidt, C. 1865. "Uber die Unechtheit des Dialogs Kratylos." Rheinisches
Museum fiir Philologie, n.s., 20: 321-56.
Schaerer, R. 1930. 'Enioir\\ir\ et Te^vi): Etude sur les notions de connaissance et d'art
d'Homere a Platen. Macon: Protat Freres.
Schaper, E. 1968. Prelude to Aesthetics. London: George Allen and Unwin.
Schmitz, H.-G. 1991. "Die Eroffnung des sprachphilosophischen Feldes: Uberlegungen
zu Platons Kratylos." Hermes 119: 43-60.
Schofield, M. 1979. Review of Irwin 1977^ Classical Review, n.s., 29: 246-49.
. 1982. "The denouement of the Cratylus." In Language and Logos, 61-81. Ed.
M. Schofield and M. C. Nussbaum. Cambridge: Cambridge University Press.
Sedley, D. 1998. "The Etymologies in Plato's Cratylus." Journal of Hellenic Studies 118:
140-54.
Selincourt, A. de, tr. 1972. Herodotus, Histories. Revised, with an introduction and notes
by A. R. Burn. Harmondsworth, Middlesex: Penguin Books.
Shorey, P., tr. [1930] 1937. Plato, Republic. Loeb Classical Library. Reprinted with revi-
sions, Cambridge, Mass.: Harvard University Press.
. 1933. What Plato Said. Chicago: University of Chicago Press.
Silverman, A. 1992. "Plato's Cratylus: The Naming of Nature and the Nature of Nam-
ing." Oxford Studies in Ancient Philosophy 10: 25-71.
 BIBLIOGRAPHY 185

Smith, J. A. 1917. "General Relative Clauses in Greek." Classical Review 31: 69-71.
Smyth, H. W., tr. 1922-1926. Aeschylus. Vols. 1-2. Appendix to vol. 2 ed. H. Lloyd-Jones.
Loeb Classical Library. Cambridge, Mass.: Harvard University Press.
Snell, B. 1926. "Die Sprache Heraklits." Hermes 61: 353-81.
. [1953] 1982. The Discovery of the Mind in Creek Philosophy and Literature. Tr.
T. G. Rosenmeyer. Reprint, New York: Dover.
Solmsen, F., ed. 1983. Hesiod, Theogonia; Opera et Dies; Scutum. Fragmenta Selecta,
ed. R. Merkelbach and M. L. West. 2d ed. Oxford: Clarendon Press.
Sprague, R. K. 1972. "Empedocles, Hera, and Cratylus 4040." Classical Review, n.s.,
22: 169.
. 1976. Plato's Philosopher-King: A Study of the Theoretical Background. Colum-
bia: University of South Carolina Press.
Stanford, W. B., ed. 1963. Aristophanes, Frogs. 2d ed. London: Macmillan.
. 1983. Greek Tragedy and the Emotions: An Introductory Study. London: Routledge
and Kegan Paul.
Steiner, A. 1916. "Die Etymologien in Platons Kratylos." Archiv fur Geschichte der
Philosophic, n.s., 22: 109-32.
Steinthal, H. [1890] 1961. Geschichte der Sprachwissenschaft bei den Griechen und
Romern. Vol. i. Reprint, Hildesheim: Georg Olms Verlagsbuchhandlung.
Strubbe, }. H. M. 1991. "'Cursed Be He That Moves My Bones.'" In Magika Hiera:
Ancient Greek Magic and Religion, 33-59. Ed. C. A. Faraone and D. Obbink. New
York: Oxford University Press.
Tambiah, S. J. 1973. "Form and Meaning of Magical Acts: A Point of View." In Modes
of Thought: Essays on Thinking in Western and Non-Western Societies, 199-229.
Ed. R. Horton and R. Finnegan. London: Faber and Faber.
Tate, ). 1928. "'Imitation' in Plato's Republic." Classical Quarterly 22: 16-23.
. 1932. "Plato and 'Imitation.'" Classical Quarterly 26: 161-69.
. 1938. "Plato, Art and Mr. Maritain." New Scholasticism 12: 107-42.
Taylor, A. E. 1960. Plato: The Man and His Work. 7th ed. London: Methuen.
Taylor, C. C. W. 1979. Review of Irwin 1977^ Mind 88: 597-99.
Thayer, H. S. 1969. "Models of Moral Concepts and Plato's Republic." Journal of the
History of Philosophy 7: 247-62.
Thesleff, H. 1982. Studies in Platonic Chronology. Commentationes Humanarum
Litterarum, vol. 70. Helsinki: Societas Scientiarum Fennica.
Tigerstedt, E. N. 1969. Plato's Idea of Poetical Inspiration. Commentationes Humanarum
Litterarum, vol. 44, no. 2. Helsinki: Societas Scientiarum Fennica.
. 1970. "Furor Poeticus: Poetic Inspiration in Greek Literature before Democritus
and Plato." Journal of the History of Ideas 31: 163-78.
Tiles, J. E. 1984. "Techne and Moral Expertise." Philosophy 59: 49-66.
Tsantsanoglou, K. 1997. "The First Columns of the Derveni Papyrus and Their Reli-
gious Significance." In Laks and Most 1997, 93-128.
Urmson, J. O. 1982. "Plato and the Poets." In Plato on Beauty, Wisdom, and the Arts,
125-36. Ed. }. Moravcsik and P. Temko. Totowa, N.J.: Rowman and Littlefield.
Vellacott, P, tr. 1972. Euripides, "Orestes" and Other Plays. London: Penguin Books.
Verdenius, W. J. 1970. "Homer, the Educator of the Greeks." Mededelingen der Koninklijke
Nederlandse Akademie van Wetenschappen, Afd. Letterkunde 33: 207-31.
. 1983. "The Principles of Greek Literary Criticism." Mnemosyne 36: 14-59.
Versnel, H. S. 1991. "Beyond Cursing: The Appeal to Justice in Judicial Prayers." In
186 BIBLIOGRAPHY

Magika Hiera: Ancient Greek Magic and Religion, 60-106. Ed. C. A. Faraone and
D. Obbink. New York: Oxford University Press.
Vlastos, G. 19653. "The Third Man Argument in the Parmenides." In Allen 196513, 231-
63. Originally published in Philosophical Review 63 (1954): 319-49.
. 1965^ "Postscript to the Third Man: A Reply to Mr. Geach." In Allen 1965^
279-91. Originally published in Philosophical Review 65 (1956): 83-94.
. 1969. "Plato's 'Third Man' Argument (Parm. i32Ai-B2): Text and Logic." Philo-
sophical Quarterly 19: 289-301.
. 1971. "Introduction: The Paradox of Socrates." In The Philosophy of Socrates: A
Collection of Critical Essays, 1-21. Ed. G. Vlastos. Notre Dame: University of Notre
Dame Press.
. 1981. "On a Proposed Redefinition of'Self-Predication' in Plato." Phronesis 26:
76-79.
. 1987. "'Separation' in Plato." Oxford Studies in Ancient Philosophy 5: 187-96.
. 19952. Review of Irwin ig77b. In Studies in Greek Philosophy, vol. 2: Socrates,
Plato, and Their Tradition, 126-32. Ed. D. W. Graham. Princeton: Princeton
University Press. Originally published in Times Literary Supplement, February 24,
1978, 230-31.
-. ig95b. "Was Plato a Feminist?" In Studies in Greek Philosophy, vol. 2: Socrates,
Plato, and Their Tradition, 133-43. Ed. D. W. Graham. Princeton: Princeton
University Press. Originally published in Times Literary Supplement, March 17,
1989, 276, 288-89.
Walzer, R. R., and J. M. Mingay. 1991. Aristotle, Ethica Eudemia. Oxford: Clarendon
Press.
Warburg, M. 1929. Zwei Fragen zum "Kratylos." Grafenhainichen: C. Schulze.
Ward,}. K. 1995. "Focal Reference in Aristotle's Account of <J>i?iia: Eudemian Ethics
VII 2." Apeiron 28: 183-205.
. 1996. "Souls and Figures: Defining the Soul in De anima ii 3." Ancient Philoso-
phy 16: 113-28.
Warren, E. 1989. "The Craft Argument: An Analogy?" In Essays in Ancient Greek Phi-
losophy, vol. 3: Plato, 101-15. Ed. J. Anton and A. Preus. Albany: State University
of New York Press.
Waterfield, R., tr. 1993. Plato, Republic. Oxford: Oxford University Press.
Webber, A. 1989. "The Hero Tells His Name: Formula and Variation in the Phaeacian
Episode of the Odyssey." Transactions of the American Philological Association 119:
1-13.
Wecklein, N., ed. 1981. Aeschylus, Prometheus Bound; fragments of Prometheus Un-
bound. Tr. F. D. Allen. New Rochelle, N.Y.: Caratzas Brothers.
Wehrli, F., ed. 1969. Die Schule des Aristoteles. Vol. 7: Herakleides Pontikos. 2d ed. Basel:
Schwabe.
Weingartner, R. H. 1973. The Unity of the Platonic Dialogue. Indianapolis: Bobbs-
Merrill.
Weiss, R. 1998. Review of Roochnik 1996. Ancient Philosophy 18: 177—82.
West, M. L. 1983. The Orphic Poems. Oxford: Clarendon Press.
. 1997. "Hocus-Pocus in East and West: Theogony, Ritual, and the Tradition of
Esoteric Commentary." In Laks and Most 1997, 81-90.
White, N. P. 1976. Plato on Knowledge and Reality. Indianapolis: Hackett.
. 1979. A Companion to Plato's "Republic." Indianapolis: Hackett.
 BIBLIOGRAPHY 187

. 1986. "The Rulers' Choice." Archiv fur Geschichte der Philosophic 68: 22-46.
Wilamowitz-Moellendorff, U. von. 1959. Platan: Sein Leben und Seine Werke. 5th ed.
Revised and provided with an afterword by B. Snell. Berlin: Weidmannsche
Verlagsbuchhandlung.
Wild, J. 1946. Plato's Theory of Man: An Introduction to the Realistic Philosophy of
Culture. Cambridge, Mass.: Harvard University Press.
Williams, B. 1982. "Cratylus" Theory of Names and Its Refutation." In Language and
Logos, 83-93. Ed. M. Schofield and M. C. Nussbaum. Cambridge: Cambridge
University Press.
. 1993. Shame and Necessity. Sather Classical Lectures, vol. 57. Berkeley: Univer-
sity of California Press.
Wilson, J. R. 1968. "The Etymology in Euripides, Troades, 13-14." American Journal of
Philology 89: 66-71.
Woodruff, P., tr. 19823. Plato, Hippias Major. Indianapolis: Hackett.
. ig82b. "What Could Go Wrong with Inspiration? Why Plato's Poets Fail." In
Plato on Beauty, Wisdom, and the Arts, 137-50. Ed. J. Moravcsik and P. Temko.
Totowa, N.J.: Rowman and Littlefield.
, tr. 1983. Plato, Two Comic Dialogues: "Ion" and "Hippias Major." Indianapolis:
Hackett.
. 1990. "Plato's Early Theory of Knowledge." In Companions to Ancient Thought,
vol. i: Epistemology, 60-84. Ed. S. Everson. Cambridge: Cambridge University
Press.
. 1998. "Plato's Use of Poetry." In Encyclopedia of Aesthetics, vol. 3, 523-27. Ed.
M. Kelly. New York: Oxford University Press.
Wiinsch, R., ed. 1897. Defixionum Tabellae Atticae. Inscriptiones Graecae, vol. 3, pt. 3.
Berlin: G. Reimer.
Wyckoff, E., tr. 1959. Euripides, The Phoenician Women. In The Complete Greek Trag-
edies. Ed. D. Grene and R. Lattimore. Euripides, vol. 5. Chicago: Chicago Uni-
versity Press.
Young, D. C. 1966. "Sophocles Ajax 516." Classical Philology 61: 249-50.
Zeyl, D. 1982. "Socratic Virtue and Happiness." Archiv fur Geschichte der Philosophic
54: 225-38.
This page intentionally left blank
 GENERAL INDEX

Achilles, 160 terminology for discussing

Admetus, 121 appropriateness in, 63
Aeschylus aletheia, 95
on Apollo, 20 allegories, 11, 69, 92, 148 n. 64, 163-64
in Aristophanes' Frogs, 10 Allen, J., 82 n. 5
in classical education, 8 Allen, R. E., 114-15
Cratylus influenced by, 44 n.io Amyntas, 102
Eumenides, 102 Anagnostopoulos, G., 90
Libation Bearers, 15 n. 3, 56, 60, 63, 73, Anaxagoras, 31 n.45, 68, 76
120 andreia, 94
Persians, 21, 25, 55 0.33, 102, 105-6 Andromache (Euripides), 22, 25
popular acclaim of, 9 Annas, J.
as source on etymology and eponymy, on civic poetry, 153 n. 78
5n-5 on conflicting views of poetry in
See also Agamemnon; Prometheus Republic, 152
Bound; Seven against Thebes; on divine inspiration of poets, 146
Suppliants n. 57, 163 n. 96
Affinity Argument, 112 on Forms as universals, 108
Agamemnon, 18, 51, 52, 56, 57 on justice and psychic harmony, 154
Agamemnon (Aeschylus) n.8o
on Apollo, 20, 55, 62 on Line and Cave as conflicting, 156
on Clytemnestra, 20 on organizing principles in artifact
on Helen, 21-23, 121~22> 13° construction, 145
on Zeus, 30, 59 on Principle of Conflict in Republic,
agathon, 94, 131 11.14 165 n. 99
Agave, 18 Antigone, 28
Aias, 15, 16 n. 7, 17 Antisthenes, 43 n. 6, 65-66
aischron, 95 Aphetae, 25
aisthesis (sense-perception), 4, 108 Aphrodite
aistheta, 112 11.32, 155 in Derveni papyrus, 77, 79 n. 92
Ajax (Sophocles), 17, 60 etymology of name as traditional, 52
Alcestis (Euripides) Euripides on name of, 18, 31 n.44, 55,
on Apollo, 29, 55 59, 101
functional terms in, 121 Hesiod on name of, 14, 15, 31 n.44, 55>
on Hades, 61 59
on right and left, 32 0.47 Plato on name of, 59

189
190 GENERAL INDEX

Apis, 103 Austin, N., 37

Apollo auxiliaries (epikouroi), 149, 157
Aeschylus on epithet "Phoebus," 25,
55 n -33 Bacchae (Euripides)
descriptive content of name of, 13, 55, on Agave, 18
62, 130 on Demeter, 30, 53, 58
and Ion, 122 on Pentheus, 17
Plato on name of, 55, 60, 62 on right and left, 32 n.48
tragedians on name of, 20, 29 Bacchus, 103
Apology (Plato), 100 n. 2, 146 n. 56 Bambrough, R., 137 n. 33
appearances Barca, 105, 106
distinction within realm of, 155 Barney, R., 61 11.41, 70
eikasia, 83 n. 9, 155-56 Battus, 28
Parmenides and Plato contrasted Baxter, T. M. S.
regarding, 107 on Cratylus' targets as second-
See also reality-appearance dichotomy leaguers, 44 n. 11
appropriateness of onomata. See orthotes on Democritus and Cratylus, 69
onomaton on Derveni papyrus, 76, 78, 79
Am, 40 on Empedocles and etymology of
Archander, 103 "Hera," 54
Arei'thous, 25 on Euthyphro and Cratylus, 66
Ares, 29, 56, 60 on Plato on Apollo, 62 n.42
arete. See virtue on Plato on mimetic language, 90
Aristoclides, 24 n.2j on Plato's etymologies in Cratylus,
Aristophanes, 9-10, 166 5°
Aristotle on techne, 82 n. 3
on Anaxagoras, 31 11.45, 68 nn. 61-62 Belfiore, E., 134 n. 24
on eponymy, no 11.28 belief
on homonumos, 112 n. 32 knowledge distinguished from, 83,
Peri Ideon, 144, 145 H°> *55
on Plato as committed to more Forms pistis as term for, 83 n. 9
than he wants, 144, 145 See also opinion
Poetics, 48 Benardete, S., 52, 94 n. 37
on Prodicus' fifty-drachma course, 46 Bestor, T. W., 114
n. 14 bird signs, 33, 34
on Protagoras on the Iliad, 48 11.17 Brandwood, L., 4 n.4
on separateness of Platonic Forms, 3, Brown, C. S., 36, 37
107 n.2i Burkert, W.
Sophistici Elenchi, 48 on Athena and Athens, 103
Arnaeus (Irus), 105 on Derveni papyrus, 76, 77, 79
Arnim, H. von, 66 on euphemism about daimon, 35
artifact production, 143-46, 149 n.57
Astyanax (Scamandrius), 26-27, 51, 52 on Zeus, 59 11.38
Athena
aegis of, 22 n. 21 calculation (logistike), no, 116
allegiance to Odysseus, 19-20, 55-56 Calvert, B., 4 n.4
appearing as son of Anchialus, 26 Calypso, 26
as eponymous deity of Athens, 102, carpentry, 143-46, 149
103, 105, 114 Cave, the, 130 n. 12, 155-57
Homer on intellectual prowess of, 19- Charmides (Plato), 135, 136 n. 31, 141
20, 55-56, 59 Cherniss, H., 112 n. 32, 113
and phronesis, 69 Chrysaor, 15
Atlantes, 104 City (Great) Dionysia, 8-9
Atreus, 18, 51, 56 Cleisthenes, 102
Atreus, House of, 30, 51-52, 59 Clytemnestra, 20, 35 0.57, 120
 GENERAL INDEX 191

cognition structure and outcome of inquiry of,

dianoia, 141, 144, 164 80-98
Divided Line classifying, 156 on techne and naming, 82-89, H°
eikasia, 83 n.g, 155-56 n. 42
noesis, 94, 95, 108, 141 Cronus, 29, 54, 60, 77, 78
phronesis, 69, 94, 95 Cronus' Hill, 103
See also belief; knowledge; opinion Cross, R. C., 137 n. 32
Collingwood, R. G., 131, 137 n. 32, 150 n. 71 Ctesippus, 17, 101
comedy Curd, P., 3 n. 3
Aristophanes' Frogs, 9-10, 166 curses (defixiones; katadesmoi), 38-41
first comic performance, 10 n. 14 custom (ethos), 89, 93, 117
influence in Athenian culture, 9 Cyclopes, 16-17
Conflict, Principle of, 165 11.99 Cypselus, 16
convention (suntheke), 89-94, 116—17
Cooke, E. F., 151 n. 72 Dahlmann, H., 66
Cornford, F. M., 100 n.2 Danto, A. C., 168-69
correctness of onomata. See orthotes defixiones (curses), 38-41
onomaton Deianira, 120-21
Country Dionysia, 9 Deio, 77, 79
craft analogy, 6, 82, 135-36, 137 n. 33 deities
Craik, E. M., 72 Ares, 29, 56, 60
Crathis, 104 Bacchus, 103
Cratylus, 43 n. 6, 45, 47 n. 16, 64-65 Cratylus on divine names, 58-62, 87,
Cratylus (Plato) 170-71
conclusion of, 94-98 in Derveni papyrus, 77, 79 n. 91
craft analogy in, 6, 82 Dike, 15, 56, 60
as crucial locus of Plato's theorizing, 98 Dionysus, 72—73
and Derveni papyrus, 76-79 and Forms, 96 n.43
on divine names, 58-62, 87, 170-71 and men using different onomata, 50
on eponymy, 98 names of, 6, 28-30, 58-62, 170-71
etymology in, 4, 6, 13, 42, 50, 71, 81 Pelasgians on, 54
n.2, 92, 118, 129-31 Plato on poetic treatment of, 6, 58-63,
Forms in, 96 n.44 151
on functional terms, 98 See also Aphrodite; Apollo; Athena;
Greek literary tradition's influence on, Cronus; Demeter; Hades; Hera;
5-6, 13, 31, 42-79, 169-70 Zeus
as intersection of philosophical and Demand, N., 85
literary traditions, 44 Demaratus, 15-16
investigating historical sources of, 43-45 Demeter
as middle dialogue, 4 n.4 in Derveni papyrus, 77, 78, 79
nature-convention dichotomy in, 89— and Earth, 30, 53, 78
94, 116-17 Plato on derivation of, 58, 78
orthotes onomaton in, 4, 42, 45-63, 71, Democritus, 4, 10, 68, 69
95 Demodocus, 19, 101
and Phaedo, 98 n.47, 116-17, u& n-42 Demonax, 28
philosophical concerns of, 6-7, 80-81 DeMundo (pseudo-Aristotelian treatise), 53
philosophy of language in, 42 Derbolav, ]., 43, 44 n.ii, 64, 65 11.47, ^6
on poetry, 134-35, H7 n. 16
and Republic, 57-62, 85, 118, 170-71 Derveni papyrus, 76-79
seriousness of, 6-7 dialectic, 89
on Socrates and Prodicus' fifty- dialektikos, 84, 86
drachma epideixis, 46, 47 dianoia, 141, 144, 164
on sophia of Socrates, 86 Diels, H., 44 n.ii
sources for discussion of etymology in, dikaion, 56, 68, 94, 116, 131 n. 14
42-79 dikaiosune, 56, 94
192 GENERAL INDEX

Dike, 15, 56, 60 eponymy

Dimock, G. E., Jr., 37 Aristotle on, no n.28
Diogenes of Apollonia, 76 in Cratylus, 98
Dionysus, 72-73 dependencies in Platonic, 99
Divided Line etymology compared with, 101-2
the Cave as conflicting with, 156 and Forms, 7, 99, 107—8, no— n
distinction within appearances in, 155 in later dialogues, 117
and Form of the Good, 140 n.^i literary, 101-6
on mathematics and dianoia, 144 literary tradition influencing Plato's
and mathematics in hierarchy of view of, 7
technai, 139-41 nonphilosophical uses of, in n. 30
pistis and eikasia as divisions of doxa and participation, 99-100, 113, 114
in, 83 n. 9 in Phaedo, 4, 7, 98, 99, 100 n. 5, 117,
divine omens, 33 n. 51 118, 125, 170
divinities. See deities in Plato, 106-17
Division, Method of, 89 n.24, 162 Plato's philosophical predecessors
Dodds, E. R., 83 n. 9 and, 100
Dolon, 18, 101 Plato's sources on, 5 11.5
Dover, K. }., 10 n. 14, 166 n. 102 and Third Man Argument, 114-15, 117
doxa. See opinion as wordplay, 71
ems, 93 n. 32, 95, r6i
Edmundson, M., 169 ethics
education influences on Plato's, 4
literature in classical, 8 moral knowledge as craft, 136 n. 32
paideia, 149, 163 See also Good, the; justice; values;
philosophy's role in, 129, 166 virtue
poetry's role in, 129, 135, 149, 152-53, etymology
163, 166 in Cratylus, 4, 6, 13, 42, 50, 71, 81 n.2,
Edwards, M. }., 77, 79 n. 93 92, 129-31
eikasia, 83 11.9, 155-56 Cratylus as etymologist, 64
eikones, 66 and Cratylus' sources, 42-79
Electra, 120 distinguishing from other forms of
Electra (Euripides), 32 n.48, 103 wordplay, 70-71
elements, 54 n. 32, 131 n. 14 episteme and Plato's approach to, 96
Elias, J. A., 92 11.31, 131 eponymy compared with, 101-2
Else, G. F., 156 n. 82, 164 n. 97 etymologizing as techne, 6, 80, 88—89
Empedocles, 54 and Forms, 130
empeiria, 82 11.4, 83 in Heraclitus, 71-75
encomia, 159-62, 164 n. 97 literary, 13-31, 80, 118, 169-70
Encomium of Helen (Gorgias), 22 11.20 nature as revealed in, 13, 56, 80, 92
Epaphus, 15 n. 30, 101, 129, 162, 170
Epigoni, 25 Plato's sources on, 5 n. 5
epikouroi (auxiliaries), 149, 157 Plato's use of outside Cratylus, 92-93
Epimetheus, 19 Prodicus as concerned with, 45
episteme Protagoras as concerned with, 45
doxa contrasted with, 95 right ontology and determinations of
as key Platonic term, 94 fitness by, 91
and phronesis and sophia, 94 n. 34 eudaimonia, 6, 81
in Plato's analysis of etymological Eumenides (Furies), 35-36, 38, 58 n. 35
approach to fitness, 96 Eumenides (Aeschylus), 102
Plato's dual etymologies of, 91 euonumos, 34
studies for generating, 163 Euripides
in techne, 83, 94 n. 35 Andromache, 22, 25
See also knowledge in Aristophanes' Frogs, 10
epithumia, 95 in classical education, 8
 GENERAL INDEX 193

Cratylus influenced by, 44 n.io Fowler, H. N., 51 0.25, 61 11.41, 81

Electra, 32 n.48, 103 Fraenkel, E., 23
Heracles, 61 Frege, G., 117
Hippolytus, 18 Friedlander, P., 49, 61 11.41
Iphigenia in Tauris, 19, 63, 101, 104 Frogs (Aristophanes), 9-10, 166
Medea, 32 functional terms
Rhesus, 18, 101 in Cratylus, 98
as source on etymology and eponymy, descriptive and prescriptive
511.5 application of, 123
Suppliants, 17 n.io, 18, 24-25 judgments of fitness involving, 71
See also Alcestis; Bacchae; Helen; Ion; in Republic, 4, 7, 98, 123-25, 170
Iphigenia in Aulis; Orestes; in tragedy, 7, 119-22
Phoenician Women; Trojan Women Funghi, M. S., 77
Euthydemus (Plato), 47 11.15, 67, 82 n.4 Furies (Eumenides), 35-36, 38, 58 11.35
Euthyphro, 43 n. 6, 66-67, 76 n. 81
Euthyphro (Plato), 67 Gaiser, K., 43, 44 n. n, 87 n. 18, 96 11.44
Euxenus, 24 11.27 Gallop, D., 112 n. 32
gods. See deities
Faraone, C. A., 38-39, 40 Goldschmidt, V., 78 11.90
Ferwerda, R., 85 0.15 Gonzalez, F, ]., 11 11.19, 92 11.30, 136 11.29
Fine, G., 145 Good, the
fine names (kala onomata), 87, 131 n. 14 Form of the Good, 138, 140
fitness of onomata. See orthotes paideia leading to apprehension of,
onomaton 163
flux, doctrine of, 64, 69-70, 76 philosophers as apprehending, 138-39,
focal meaning, no 11.28 140
Fogelin, R. }., 140 n.4i and philosophy as techne, 128
Forms Gorgias, 22 n. 20
as being unqualifiedly what they are, Gorgias (Plato)
125 on cookery as empeiria, 82 11.4
in Cratylus, 96 n. 44 Cratylus as linked to, 82
dianoia as directed to, 141, 144 on criteria for techne, 80, 83, 128, 142
and divine individuals, 96 n.43 and the early dialogues, 84 n.n
and divine names, 63 and goodness, 128 n. 5
eponymy and, 7, 99, 107-8, 110-11 on goodness and order, 142 n.45
and etymology, 130 knowledge distinguished from
explanatory function of, 106-7, U5 opinion in, 83 n. 9, 146 n. 55
Form of the Good, 138, 140 myth at end of, 92
having their own onomata applied to on poetry as techne, 132
them, 111 on rhetoric as techne, 6, 80, 81
hierarchy of, 138, 144 sdma-psuche distinction in, 95
influences on Plato's theory of, 3, 100 techne and empeiria distinguished in,
in later dialogues, 117-18 82 11.4, 83
mathematics as concerned with, 109, Gottschalk, H. B., 66
!?9 Gould, ]., 136
and nature, 96, 109-10, 115-16, 124 Great (City) Dionysia, 8-9
and onoma construction, 88 Greek literary tradition
philosophers knowing, 125, 138, 140 in classical education, 8
poetry criticized in light of, 155 Cratylus as intersection of
primacy of, in, 130 n. 12 philosophical tradition and, 44
as "role models," 151 Cratylus influenced by, 5-6, 11, 13, 42-
separateness of, 107 n.2i 79
and techne, 83 eponymy in, 101-6
types of characteristics of, 115 etymology in, 13-31, So, 118, 169-70
See also participation on functional terms, 119-22
194 GENERAL INDEX

Greek literary tradition (continued) Heracles (Euripides), 61

individual natures as represented in, Heraclides Ponticus, 43 n. 6, 45 n. n, 66
30 11.43, 80 n.i, 129 Heraclides Ponticus the Younger, 66
on natural correctness of words, 49-50, Heraclitus
0.
°4 Cratylus influenced by, 5, 43 n. 6, 68,
Plato as influenced by, 4, 8-11, 169 75-76
and Plato's accounts of others' views, and Derveni papyrus, 76
67-68 on drunkenness, 73 0.72
Plato's hostility toward, 49 and eponymy, 100
Plato's view of eponymy influenced by, 7 flux doctrine of, 64, 69-70, 76
Republic distinguishing literature Plato's ethics influenced by, 4
from philosophy, 11 theory of Forms influenced by, 3, 100
terminology for discussing wordplay in, 71-75
appropriateness in, 63 Hermogenes, 47, 48
See also Herodotus; poetry; tragedy Herodotus
Greene, W. C., 148 11.64 on Amyntas, 102
Grene, D., 34 n. 55 on Aphetae, 25
Grey, D. R., 131 n. 15 on Archander, 103
grief, 165 on Atlantes, 104
Griffith, M., 29 n. 39 on Barca, 105
Griswold, C., 147 on Battus, 28
Grote, G., 45 on Cleisthenes, 102
Grube, G. M. A., 140 n.42 on Crathis, 104
guardians, 123, 124, 132, 151 nn. 71-72 Cratylus influenced by, 44 n.io
Guthrie, W. K. C., 45, 46 nn. 13-14, 50, on Demaratus, 15-16
54 n. 32, 65 on Demonax, 28
Gygean Treasure, 103 on Egyptian religion, 36
on Gygean Treasure, 103
Hades on Hegesistratus, 28
Heraclitus on Dionysus and, 72, 73 on Ion and the lonians, 102-3
Homer on name of, 29, 53 on the Melanchlaeni, 26
Plato on fear generated by name of, on Miltiades, 102
53, 60-61 on Mother of Hypanis, 105
Plato's differing analyses of, 93 on the Pelasgians, 30, 54
Hall, R. W., 84 n.n, 128, 137 n.32 on Pisistratus, 102
Halliwell, S., 147 n. 59, 158 on right and left, 32, 34 11.53, 35
Hartland-Swann, ]., 131-32 on Samius, 104
Heath, D. D., 64-65, 70 n.65 on the Scoloti, 102
heavenly bodies, 54 as source on etymology and eponymy,
Hector, 13, 26-27, 51, 52, 56 5 n.5
hedone, 95, 132 terminology for discussing
Hegesistratus, 28 appropriateness in, 63
Helen, 21-23, 25, 121-22, 130 on three names for one landmass, 105
Helen (Euripides) heroes, 56-57, 132, 151
on Demeter, 30, 53, 58, 78 Hesiod
on Helen, 25 in classical education, 8
on Hera, 29, 54 on the Cyclopes, 16-17
on Ido (Theonoe), 18-19, 101 date of, 5 n.5
on New Salamis, 104 and Derveni papyrus, 79
on right and left, 32 nn. 46, 48 and Plato's etymology of da.im.on, 68
Hera See also Theogony; Works and Days
in Derveni papyrus, 77 Hippasus, 18
Euripides on, 29, 54 Hippias Major (Plato), 117 n. 38
and the four elements, 54 n. 32 Hippias Minor (Plato), 142
Iliad on, 29, 54, 59, 70 Hippolytus (Euripides), 18
 GENERAL INDEX 195

Hippomedon, 18 on Athena's aegis, 22 n.2i

Hirschle, M., 69 n. 64 functional terms in, 122
Hogan, R., 136 n. 31 on Ion and the lonians, 101, 102-3
Homer Ion (Plato)
as banned from the Republic, 49 on poetry as techne, 81, 82 n. 7, 134
in classical education, 8 n. 21
date of, 5 n. 5 on poets as inspired, 146 n. 56, 147
and Euthyphro, 67 on poets as lacking knowledge, 82
Ion on divine inspiration of, 147 n.y, 134
Plato on learning from, 49, 50-51, 169 Iphigenia in Aulis (Euripides), 18, 32
Plato's preferred poetry contrasted nn.46-48, 56, 57
with, 159 Iphigenia in Tauris (Euripides), 19, 63,
in Plato's treatment of character, 4 101, 104
on Scylla, 16 Irus (Arnaeus), 105
on the soul, 68 Irwin, T. H., 135 n. 27, 136 nn. 29, 32
as tragedian to Plato, 130 n. 8 Isocrates, 8 n. 9
See also Iliad; Odyssey
homonumos, 112-13 Jaeger, W., 10
homonymy, no Janaway, C., 131 n. 15, 134 n. 21, 153 n. 78,
House of Atreus, 30, 51-52, 59 158 n.8$
Hybristes, 22 n. 21 Jevons, F. B., 40, 41
Hyllus, 120, 122 Jordan, D., 38 n. 65, 41 0.96
hymns (humnoi), 159, 161-62, 164 0.97 Jowett, B., 46, 49 0.21, 65
Hymn to Demeter, 61 justice
etymology of "justice," 14-15, 56
lamus, 16, 31 11.44 and euonumos, 34 n. 54
ideal polls Heraclitus on, 74
auxiliaries, 149, 157 and psychic harmony, 154
guardians, 123, 124, 132, I5in n. 71-72 as techne, 137 n. 32
kinship ties in, 123-25
poets' role in, 150-67, 170 Kahn, C. H.
producer class, 151 n. 71, 164 on Democritus and Cratylus, 69
See also philosopher-rulers on Derveni papyrus, 76 n. 81, 78
Ido (Theonoe), 18-19, lo1 on Euthyphro, 67
Iliad (Homer) on Form of Beauty and a beautiful
aidos and aidoia in, 72 face, 97 n.45
on Arei'thous, 25 on the Good as focus of philosopher-
on Astyanax, 26-27, 51 rulers, 139 n. 38
in classical education, 8 on Heraclitus on great men, 74
Cratylus influenced by, 44 n.io on Heraclitus on reproduction, 73
on Hades, 29, 53 n
-73
on Hector, 26-27 on justice, 74
on Hera, 29, 54 on Parmenides' influence on theory of
on Hippasus, 18 Forms, 3 n. 3
praise of Achilles in, 160 on participation and Cratylus, 98
Protagoras' criticism of, 48 n-47
on right and left, 32, 33, 34 on Plato on opinion in naming, 87
on Simoeisius, 104 n.i8
on Thersites, 17 on Plato's middle dialogues, 3 n.i, 4
Ilium, 36 n
-4
lo, 26, 103 on proper names in Cratylus, 50
Ion, 101-3, 122 n.23
Ion (Euripides) on unity of Plato's thought, 137 n. 33
on Athena as eponymous deity of kakia, 95
Athens, 102, 103, 105 kala onomata (fine names), 87, 131 n. 14
196 GENERAL INDEX

kalon logistike (calculation), no, 116

aischron contrasted with, 95 logos, 87, 88, 92, 151
application to Form, 115, 116 Lorenz, K., 81 n.2
as key Platonic term, 94 Luce, J. V., 4 n.4, 96 n.44
in Plato's preferred poetry, 159
priority of, 131 0.14 Malcolm, J., 114 n. 35
qualified versus unqualified Marcovich, M., 70, 72, 74, 75
application of, 117 n. 38 Marrou, H.-L, 8
katadesmoi (curses), 38-41 mathematics
Kato, M., 82 n. 3, 152 0.77 dianoia as cognitive state in, 144
Kerferd, G. B, 45, 49 dual use of terms for numbers, 117
kinship Forms as concern of, 109, 139
in Plato's Republic, 123-25 and knowledge-belief distinction, 140
in tragedy, 7, 119-22 onomata of numbers, 89, 90
Kirk, G. S., 8 n. 6, 17 11.12, 70, 75 participation and arithmetic
knowledge operations, 107
belief distinguished from, 83, 140, 155 Plato on objective standards for, 3
Cratylus on, 97 as techne, 139, 141
Gorgias distinguishing opinion from, McCabe, M. M., 92 n.3i
83 n. 9, 146 n. 55 McKirahan, R. D., Jr., 100 n.2
Plato and Parmenides contrasted Medea (Euripides), 32
regarding, 4 Melanchlaeni, 26
poetry as not yielding, 49 Menelaus, 52, 57
poets as lacking, 133, 134, 147, 148-50 Meno (Plato), 137 n. 32, 144 n.49, 146
recollection theory, 137 n. 32 11.56
and techne, 83, 85, 94 n. 35, 147 Menoeceus, 102
See also episteme Meridier, L., 45 n.n, 51 n.25, 64, 93
Kranz, W., 29 n. 39 Merkelbach, R., 77 n. 85
Kraut, R., 139 n. 37 Method of Division, 89 n.24, 162
Kretzmann, N., 84 n. 12 Miltiades, 102
Kube, J., 136 n.ig mimesis
as banned from the ideal state, 153
Laches (Plato), 135, 141 and the Cave, 157
Laks, A., 76 good and bad senses of, 158
language Plato on mimetic language, 90
in curses, 40 and Plato's opposition to poetry, 127
Plato's philosophy of, 4-5, 42 in poetry for children, 152
and reality, 4-5, 13, 28, 30, 31-41, 45, Mittelstrass, J., 81 n.2
74.77 Moira, 77
See also onomata Moore, G. E., 108, 109
Lattimore, R., 166 n. 102 Moravcsik, J.
Laws (Plato), 8, 118 n.4i on Forms and techne, 83 n. 8
left and right, 31-35 on Forms' scope, 108
Lenaia, 9, 72, 73 on functional specifications in artifact
Lenardon, R. J., 69 production, 88 n. 21, 145
Libation Bearers (Aeschylus) on goodness condition for techne, 83
on Dike, 15 n. 3, 56, 60 on productive model in Greek
on functional terms, 120 thought, 106 0.19
on Moira and moms, 73 on self-exemplification, 114 n. 35
terminology for discussing on types of characteristics of Forms,
appropriateness in, 63 n
5
Line. See Divided Line Morford, M. P. O., 69
literary tradition. See Greek literary Most, G. W., 76
tradition Mother of Hypanis, 105
Lodge, R. C., 147 n. 59, 148 0.64 Murdoch, I., 166 n. 102, 168, 171
 GENERAL INDEX 197

Murphy, N. R., 168 Athena appearing as son of Anchialus

Murray, G., 166 11.102 in, 26
Murray, P., 146 nn. 56, 58, 153, 157, 160 on Calypso, 26
n. 90, 164 0.97 Cratylus influenced by, 44 n.io
myths, 11, 69, 92, 148 n. 64, 163-64 on Ctesippus, 17, 101
on Demodocus, 19, 101
Nagy, G., 35 n.57 on gates of horn or ivory, 25
names. See onomata on Irus (Arnaeus), 105
natures (phuseis) on Neoptolemus, 24
Antisthenes and natural-correctness Odysseus' name suppressed in, 37-38
theory, 65 on Penelope, 25, 36
as catch-word of Greek reflection, 46 on right and left, 32 11.47, 33> 34
n.i3 on Styx, 58 0.35
and convention, 89-94, 116-17 Oedipus, 16, 28
Cratylus as concerned with, 6 Oedipus at Co/onus (Sophocles)
Democritus and natural-correctness on Antigone, 28
thesis, 69 on Athena and Athens, 102, 103, 105
and divine names, 62-63 on functional terms, 122
etymology as revealing, 13, 56, 80, 92 on the Furies, 35
n. 30, 101, 129, 162, 170 on Parthenopaeus, 15
and Forms, 96, 109-10, 115-16, 124 on right and left, 32
of individuals, 115, 116, 129 Oedipus Tyrannus (Sophocles), 16, 29,
literary tradition on individual, 30 55- 103
n.43, 80 n.i, 129 omens, 33 n. 51
literary tradition on natural onomata
correctness, 49-50, 84 belief in ties with reality of, 31-41
ousia, 84, 86, 95, 96, no criteria for assigning, 14
in Phaedo, 109 fine names (kala onomata), 87, 131 11.14
Plato on natural-correctness doctrine, gods and men using different, 50
49, 81, 84-85, 89, 91, 92-93 naming as techne, 84-89
Plato on poets' misunderstanding of, and natures as philosophers'
!33> >-34> l62 responsibility, 162
and Plato's analysis of dikaion, 56 nicknames, 105
as unchanging, 96, no notions included in concept of, 13 n.i
Nehamas, A., 152 n. 77, 153 n. 78, 158 of numbers, 89, 90
Neoptolemus, 24 Plato addressing topic of names, 6
Nettleship, R. L., 152, 156 Plato's key philosophical terms, 94-95
New Salamis, 104 proper names, 13, 50 n.23, 80, 130 11.9
nicknames, 105 Republic on descriptive content of, 57
Nightingale, A. W., 160 shifts for apotropaic purposes, 34-38
Nisaean horses, 104 See also eponymy; etymology;
Nisetich, F. J., 24 n. 27 functional terms; orthotes
noesis, 94, 95, 108, 141 onomaton
nomos, 46 n. 13, 133 On Truth (Protagoras), 49
nomothetes, 84, 85-86 opinion (doxa)
numbers, 89, 90, 117 in artifact production, 145
Nussbaum, M. C., 68 n.6o and dianoia, 141
Gorgias distinguishing knowledge
oaths, 31-32 from, 83 n. 9, 146 11.55
O'Brien, M. )., 136 n. 29 Plato on naming and, 86-87
Odysseus, 18, 19-20, 37-38, 55-56 Plato on status of, 4
Odyssey (Homer) as poets' level of cognition, 133, 134, 157
on Aias, 17 poets requiring true, 149
on Ares, 29 n. 38 terms contrasted with, 95
on Athena and Odysseus, 19-20, 55-56 the young absorbing from poets, 150
198 GENERAL INDEX

Opous, 102 in later dialogues, 117-18

opposites, 99, 109 participants' being what they are only
Orestes, 51, 103 qualifiedly, 115, 125
Orestes (Euripides) Pegasus, 15
on Apollo, 20, 55 Peitho, 79 n.ga
on Clytemnestra as mother, 120 n.48 Pelasgians, 30, 54, 102, 105
trepidation about naming entities in, Pelasgus, 14, 102, 105
36 Pelops, 51
on Zeus, 30 Penelope, 25, 36
Orion, 66 Pentheus, 17, 130
Orphism, 76, 77, 85 Peradotto, J., 26 n. 29
orthotes onomaton perception (aisthesis), 4, 108
Antisthenes and natural-correctness Peri Ideon (Aristotle), 144, 145
theory, 65 Persians (Aeschylus), 21, 25, 55 0.33, 102,
and dating of Cratylus, 4 n.4 105-6
conceptions of appropriateness, 99- Pfeiffer, R., 44 n.n, 82 n. 3
126 Phaedo (Plato)
in Cratylus, 4, 42, 45-63, 71, 95 on Anaxagoras, 68
Democritus and natural-correctness central cases of Forms in, 144
thesis, 69 and Cratylus, 98 n.47, 116-17, u$ n-42
divine provenance in, 23-24 dating of, 4 n.4, 118 n.42
eponymy and, 105-6 on deficiency of particulars, 107-8
of functional terms, 119-25 on eponymy, 4, 7, 98, 99, 100 n. 5, 117,
literary tradition influencing Plato's 118, 125, 170
handling of, 5-6, 11, 169 on fire and snow, 96 n.43, 109 11.25
literary tradition on natural on Forms' and understanding, 148
correctness, 49-50, 84 on Forms' being unqualifiedly what
multiple constructions of, 13 they are, 115
and nature-convention distinction, on Forms having their own onomata
89-94, 116-17 applied to them, 111
participation and Cratylus' treatment Forms' scope in, 108-9
of, 97 on Hades, 61 11.41, 93
Plato on natural-correctness doctrine, as middle dialogue, 3 n.i, 4 n.4
49, 81, 84-85, 89, 91, 92-93 myth at end of, 92
Plato's approach to, 97-98 on nature, 109
terminology for discussing on participation, 99, 107, 115
appropriateness, 63 reality-appearance dichotomy in, 96
Osiris, 36 and Republic 7, 118
ousia, 84, 86, 95, 96, no soma-psuche distinction in, 95
Owen, G. E. L., 113 on the spatiotemporal realm as
secondary, 130 n. 12
paideia, 149, 163 twofold application of onomata in, 98
Pandora, 21 Phaedrus (Plato), 50, 118 n.42, 146 n. 56, 171
Parmenides Philoctetes (Sophocles), 17-18, 21, 24, 74,
on appearances, 107 101
and eponymy, 100 philosopher-rulers
on knowledge, 4 the Good as apprehended by, 138-39,
and Republic, 10 n. 17 140
on right and left, 31 n.45 philodoxoi contrasted with, 125
theory of Forms influenced by, 3, 100 poets as auxiliaries to, 163, 164
Parmenides (Plato), 108 n. 22, 117 as ruling and philosophizing, 139 n. 37
Parthenopaeus, 15, 17 and statesmanship as techne, 137 n. 33
participation, 107 philosophy
and Cratylus, 97-98 Cratylus as intersection of literary
and eponymy, 99-100, 112, 114 tradition and, 44
 GENERAL INDEX 199

Cratylus' philosophical import, 6-7, and curses, 38 n. 65

80-81 and Derveni papyrus, 76
Forms as known by philosophers, 125, eponymy and its metaphysical
138, 140 background in, 106-17
onomata as connected to nature as hostility toward literary tradition, 49
responsibility of, 162 key philosophical terms of, 94-95
pedagogical function of, 129, 166 on learning from Homer, 49
poetry and rhetoric contrasted with, literary tradition influencing, 8-11, 169
80-81 middle dialogues of, 3-5
poetry's relationship to, 143, 148-50, myths in dialogues of, n, 92
162-64 on natural-correctness doctrine, 49,
quarrel with poetry, 129, 168-69 81, 84-85, 89, 91, 92-93
Republic distinguishing literature philosophy of language in, 4-5, 42
from, 11 on poetry (see poetry)
as techne, 12, 81, 128, 135-43 sophists sharing common literary
See also philosopher-rulers; heritage with, 46
Presocratics and wordplay, 71
Phoebus, 102 See also Cratylus; Gorgias; Ion;
Phoenician Women (Euripides) Phaedo; Republic; Sophist;
on Demeter, 30, 53, 58 Symposium; and other dialogues by
on Menoeceus, 102 name
on Oedipus, 16 Plotinus, 148 n. 64
on Polynices, 20-21, 24, 109 n.27 Plutus, 61
terminology for discussing Poetics (Aristotle), 48
appropriateness in, 63 poetry
phronesis, 69, 94, 95 carpenters compared with poets, 149
phuseis. See natures civic role of, 153
Pickard-Cambridge, A., 9 n.io criticism of Plato's view of, 168-69, 171
Pindar deities as depicted in, 6, 58-63, 151
on Aias, 15 divine inspiration of poets, 146-48, 163
on Aristoclides, 24 n.27 0.96, 171
on Athena, 20, 56 eikasia as condition of poets, 155-56
on city called after Mount Aetna, 104 emotion in, 57-58, 130 n.n, 165
in classical education, 8 human beings depicted in, 158 n, 88
Cratylus influenced by, 44 n.io "hymns to the gods and praises of
on Cronus' Hill, 103 good people," 152-53, 158-62
on Epimetheus, 19 knowledge as lacking in poets, 133,
on eponymous nymphs, 103 134, 147, 148-50
euphemism in, 35 n. 57 nature as misunderstood in, 133, 134, 162
on Euxenus, 24 n.27 opinion as poets' level of cognition,
on lamus, 16, 31 n.44 !33> 134. 157
on Opous, 102 pedagogical function of, 129, 135, 149,
on Prophasis, 19 152-53, 163, 166
as source on etymology and eponymy, Plato as influenced by, 8-11
5n-5 Plato contrasts philosophy with, 81
on Strepsiades, 102, 112 Plato on knowledge as not yielded by,
and wordplay in Heraclitus, 72 49
xenia in odes of, 24 n.27 Plato on philosophy's relationship to,
on Zeus, 30 143, 148-50
Pisistratus, 102 in Plato's ideal polis, 150-67, 170
pistis, 83 n. 9 poets engaging in etymology, 44 n.n
Plato Protagoras on importance of, 48 n. 17
on Cratylus as source for etymology, 64 quarrel with philosophy, 129, 168-69
criticism of view of poetry of, 168-69, relationship to philosophy, 143, 148-
171 50, 162-64
2OO GENERAL INDEX

poetry (continued) reality

in Republic, 127-67 Form of the Good as highest element
Republic on acceptable, 6, 57-62, 129, of, 138
143-50 language and, 4-5, 13, 28, 30, 31-41,
Republic's two discussions of, 152-66 45> ; 74>77
the soul affected by, 130, 164-65 Plato's conception of, 3-4
as techne, 6, 81, 128-35, H2> HS"?0* primary and secondary entities, 110-11
*54 See also natures; reality-appearance
See also Hesiod; Homer; Pindar dichotomy
Politicus (Plato), 89 n. 24, 98 0.49, 162 reality-appearance dichotomy
Polynices, 20-21, 24, 28, 109 n. 27, 122 in Cratylus, 97
Presocratics in metaphysics of the middle
Anaxagoras, 31 n.45, 68, 76 dialogues, 106-7, 112
Democritus, 4, 10, 68, 69 and nature, 96
Empedocles, 54 poets preferring appearance, 133
See also Heraclitus; Parmenides recollection theory, 137 n. 32
Principle of Conflict, 165 n. 99 Reeve, C. D. C., 45 n. n, 51 n. 25, 137
Principle of Specialization, 148 n. 33, 159 n. 88
Proclus, 69 relativism, 4
Prodicus Republic (Plato)
Cratylus influenced by, 5, 43 n. 6 on acceptable poetry, 6, 57-62, 129,
Derveni papyrus contrasted with, 143-50
77 the Cave, 130 n. 12, 155-57
and etymology, 45 and Cratylus, 57-62, 85, 118, 170-71
fifty-drachma epideixis of, 46, 47, 65 on deficiency of particulars, 108
orthotes onomaton as concern of, 46, epithumia-thumos distinction in, 95
47 on Forms and understanding, 148
producer class, 151 n. 71, 164 Forms' scope in, 108-9
Prometheus, 19, 47 n. 16, 82 n.4, 101 on functional terms, 4, 7, 98, 123-25, 170
Prometheus Bound (Aeschylus) and Gorgias on goodness, 128 n. 5
apotropaic shift in, 34 on "hymns to the gods and praises of
Cratus on Prometheus' name, 19, 47 good people," 152-53, 158-62
n. 16 and inspiration of poets, 147
on Cronus, 29, 54 on literature in Greek education, 8
on Hades, 61 as middle dialogue, 3 n.i, 4 n.4
on the Hybristes deserving its name, myth at end of, 92
22 n. 21 on nature and Forms, 109-10, 116
on lo, 26, 103 Parmenides' influence on, 10 11.17
on techne, 82 n.4 philosophy and literature
proper names, 13, 50 n. 23, 80, 130 11.9 distinguished in, 11
Prophasis, 19 on philosophy as techne, 135-43
Protagoras on poetry's function in the ideal polls,
Cratylus influenced by, 5, 43 n. 6 150-67
and etymology, 45 poetry's role in, 127-67
"man-measure" doctrine of, 48 on poetry's techne status, 131-35
and natural-correctness doctrine, 49-50 polls etymology in, 93
On Truth, 49 Principle of Conflict in, 165 11.99
orthotes onomaton as concern of, 46, Principle of Specialization in, 148
47-48 reality-appearance dichotomy in, 96
on poetry, 48 n. 17 on the spatiotemporal realm as
Protagoras (Plato), 47, 49, 81, 82 11.4 secondary, 130 n. 12
pseudos, 95 on techne, 81, 83 n. 7
psuche. See soul two discussions of poetry in, 152-66
punning, 70 on women, 140 n. 42
Pythagorean Table of Opposites, 31 n.45 See also Divided Line
 GENERAL INDEX 201

Rhesus (Euripides), 18, 101 sophia, 69, 86, 94, 95

rhetoric Sophist (Plato)
Plato contrasts philosophy with, 81 and Cratylus on naming, 98 nn.47,
of praise, 160 49
as techne, 6, 80, 82 n.4, 84, 86, 87, 88 on hierarchy of Forms, 138 0.36
n. 20, 129 n. 8 Method of Division in, 89 n.24, 162
Rich, A. N. M., 148 11.64 on participation and abstract entities,
Richardson, N. }., 72 11.70 118 11.41
right and left, 31-35 and Republic 10 on poetry, 156 n. 82
Rijlaarsdam, J. C., 64 and techne, 131 n. 15
Robinson, R., 13 n.i, 46 11.13, *39 n-4°> Sophistic! Elenchi (Aristotle), 48
140 n. 41 sophists
Robinson, T. M., 73 and Cratylus' handling of orthotes
role models, 151 onomaton, 5-6, 45, 49, 71
Roochnik, D. L., 136 nn.29-3i, 137 Plato's engagement with, 10
nn. 33-34 Plato sharing common literary
Ross, D., 139 n.4o heritage with, 46
Rowe, C., 152 11.76 Plato's opposition to sophistry, 3, 81
Russell, B., 108, 109 See also Prodicus; Protagoras
Rusten, }. S., 78 Sophocles
A/ax, 17, 60
Sambursky, S., 69 in classical education, 8
Samius, 104, 114 Cratylus influenced by, 44 n.io
Sansone, D., 56 Oedipus Tyrannus, 16, 29, 55, 103
Scamandrius (Astyanax), 26-27, 51, 52 Philoctetes, 17-18, 21, 24, 74, 101
Schaarschmidt, C., 97 11.46 as source on etymology and eponymy,
Schaper, E., 131 n. 15, 168, 171 n. 3 5n-5
Schofield, M., 81 11.2, 90, 91 Women ofTrachis, 32 0.46, 35, 120, 122
Scoloti, 102 See also Oedipus at Co/onus
Scylla, 16 sophrosune, 94, 132
Sedley, D., 91 11.29 soul (psuche)
selene, 68 Homer on, 68
self-attribution (self-exemplification), 114 justice and psychic harmony, 154
Selincourt, A. de, 36 n. 60 Plato's first etymology of psuche, 68
sense-perception (aisthesis), 4, 108 poetry's effect on, 130, 164-65
Seven against Thebes (Aeschylus) psuche as key term in Plato, 94
on Ares, 29, 56 soma distinguished from, 95
on Dike, 15, 56, 60 tripartite division of, 154-55
on Parthenopaeus, 17 Specialization, Principle of, 148
on Polynices, 20, 28 Sprague, R. K., 54 n. 32, 138 n. 35, 139
on right and left, 35 n.38
terminology for discussing Stanford, W. B., 130 11.9, 166 n. 102
appropriateness in, 63 Steiner, A., 65
Shorey, P., 51 0.25 Steinthal, H., 44 n. n, 64, 65, 81 11.2, 97
Silverman, A., 48 n. 19, 94 n. 33, 97-98 11.46
Simoeisius, 104 Strepsiades, 102, 112
sklerotes, 89-90, 93-94 Strubbe, J. H. M., 40
Smith, J. A., 108 n. 24 Suppliants (Aeschylus)
Smyth, H. W., 120 11.46 on Apis, 103
Snell, B., 68 n. 60, 74, 76 n. 79 on Epaphus, 15
Socrates on Pelasgus, 14, 102, 105
and eponymy, 100 on right and left, 35
Plato's ethics influenced by, 4 terminology for discussing
theory of Forms influenced by, 3, 100 appropriateness in, 63
soma, 85, 95 Suppliants (Euripides), 17 n.io, 18, 24-25
202 GENERAL INDEX

Symposium (Plato) Theonoe (Ido), 18-19, 101

on cookery as techne, 82 0.4 Thersites, 17
on deficiency of particulars, 108 Third Man Argument (TMA), 114-15, 117
on Forms and understanding, 148 Thoas, 19, 101
on Forms' being unqualifiedly what thumos, 95
they are, 115 Tigerstedt, E. N., 146-47
Forms' scope in, 108-9 Timaeus (Plato), 96 n.43, 109 n. 25, 117,
kalon-aischron distinction in, 95 137 n -33
list of aretdi in, 94 n. 36 Titans, 21
as middle dialogue, 3 n.i, 4 11.4 tragedy
on nature and Forms, 109 on Apollo, 20, 29
reality-appearance dichotomy in, 96 emotional impact of, 130 11.9
Socrates' encomium to ems in, 161 first tragic performance, 9 n.io
on the spatiotemporal realm as on functional terms, 7, 119-22
secondary, 130 n. 12 and Heraclitus on names, 70
synonymy, 71, no House of Atreus in, 52
influence in Athenian culture, 9
Tambiah, S. }., 40 n. 91 in Plato's treatment of character, 4
Tantalus, 52 as techne, 82 n.4, 129 n. 8
Tate, }., 148, 158 Zeus not presented on the stage, 59
Taylor, A. E., 46, 81 n.38
techne See also Aeschylus; Euripides;
and arete, 135-36, 141-42 Sophocles
artifact production, 143-46, 149 Trojan Women (Euripides)
in Cratylus, 82-89 on Aphrodite, 18, 31 n.44, 55> 1O1
criteria for, 80, 83, 85, 128, 132, 142 on Hector and Astyanax, 27
empeiria distinguished from, 82 n.4, on Helen as Zeus's daughter, 122
Si
8 terminology for discussing
3
etymologizing as, 6, 80, 88-89 appropriateness in, 63
and eudaimonia, 6 Tros, 102
hierarchy of technai, 139 Tsantsanoglou, K., 76 n. 84, 77, 79 n. 91
justice as, 137 n. 32
as key Platonic term, 94 Uranus, 77
and knowledge, 83, 85, 94 n. 35, 147 Urmson, J. O., 152 n. 76
mathematics as, 139, 141
naming as, 84-89 values
philosophy as, 12, 81, 128, 135-43 conventions of the many on, 155
Plato's use of term, 82 Forms pertaining to, 63, 90, 109, 144
poetry as, 6, 81, 128-35, 142> HS"?0' objective standards for, 3
!54 philosophy and, 81, 138
rhetoric as, 6, 80, 82 11.4, 84, 86, 87, 88 Plato's concern with literature and, 4
n. 20, 129 n. 8 poets and, 128, 133, 148, 150, 154
tragedy as, 82 n.4 virtue (arete)
Thayer, H. S., 137 11.33 arete as key Platonic term, 94
Theaetetus (Plato), 4 n.4, 92, 98 11.47 kakia-arete contrast, 95
Theoclymenus, 27, 33 n. 51 and techne, 135-36, 141-42
Theogony (Hesiod) Vlastos, G., 124 n. 58, 145 nn.52, 54
on Aphrodite, 14, 15, 31 n.44, 55 "voodoo dolls," 41 n. 96
on Athena, 20, 56
Cratylus influenced by, 44 n.io Warburg, M., 45 n. n, 66
on Plutus, 61 Warren, E., 135 11.27, X37 n-33
on Prometheus, 19, 101 Waterfield, R., 57 n. 35
on Styx, 58 n. 35 Webber, A., 37 n.63
succession among the gods in, 79 11.93 Wecklein, N., 29 n. 39
on the Titans, 21 Wehrli, F., 66
 GENERAL INDEX 203

Weingartner, R. H., 45, 49 on Pandora, 21

West, S., 25 n.2g on Prometheus, 19
White, N. P., 118 11.42, 164 n.g8 on Zeus, 29-30
Wilamowitz-Moellendorff, U. von, 64
Williams, B., 50 n. 23 xenia, 24 n. 27
women, 124, 140 n.42 Xerxes, 105-6
Women ofTrachis (Sophocles), 32 n.46,
35, 120, 122 Zeno of Citium, 53
Woodruff, P., 83 n.io, 137 n. 32, 143 n.47, Zeus
146, 147 in Derveni papyrus, 77
Woozley, A. D., 137 n. 32 naming of Helen by, 22-23
wordplay, 70-71 Plato on name of, 59-60
words. See onomata poetic depictions of, 151
Works and Days (Hesiod) tragedians not presented on the stage,
Cratylus influenced by, 44 n.io 59 n. 38
on Dike, 15, 56, 60 two forms of name of, 29-30, 52-53,
on Epimetheus, 19 59-60
 INDEX LOCORUM

Aeschylus 806 61 frag. 40 (Caizzi) 66

Agamemnon 840-41 103, 113 11.50
160-75 30 11.41, 53 850-51 15
n.28 910-11 29, 54 Aristophanes
681-92 21-22, 23, 122, Seven against Thebes Clouds
130
532-37 17 658-91 48
1080-82 2O, 55, 62, 577-79 20, 7111.67 1054-55 J66
130 n. g 658 20, 71 0.67
1085-86 20, 55, 130 Frogs
662-73 J 5> 56> 60, 63 1482-84 10 n. 16
11.9 829-31 20 0.19, 63
1500-1503 10 n. 16
1100-1103 20 887-90 35
1485-87 30, 52, 59 944-46 29, 56
Aristotle
Eumenides Suppliants
8 102, 105 n. 15 De Anima
40-48 15 4O4bi-6 68 n. 61
Libation Bearers 191-93 35 405313-19 68 nn. 61-62
189-91 120 250-53 14, 102, 105, 2.3 no n. 28
911 73 114
948-51 14-15, 56, 60, 260-61 103, 113 De Caelo
63, 71 0.67 3*3 J5 270)324-25 68
3!5 !5> 63 Eudemian Ethics
Persians 535 15 11.4
80 102, 105-6 584-85 30, 53 1236316-22 no n. 28
146 102 n. 8 607 32 0.46 Generation of Animals
171 21
•j6T,b^i-j6/^ai 31 n.45
177-78 21 Fragments 772b8-io 119 n.43
205-6 25, 55 11.33 35oN2 62
354 23 History of Animals
584336-1315 119 n.43
Prometheus Bound Anaxagoras
85-87 19, 44 n. n, 47 DK 11-14 68 110.61-62 Metaphysics
n. 16, 71 n. 67, 101 DK 18 68 986322-26 31 11.45
299-300 103 98713 3 n.2, 107 n.2i,
442-506 82 11.4 Antisthenes 112 n. 32
488-90 34 frag. 38 (Caizzi) 65 0.48 1003333-136 no n. 28
717—21 22 11.21 frag. 3gA (Caizzi) 65 Z.17 141 n.42
732-34 26, 71 0.67 11.49 1030229—30 no n. 28

204
 INDEX LOCORUM 205

io4ibn-33 141 0.42 77 39 n-?6> 4° 0.81 Derveoi Papyrus

10431223-32 65 79 39 nn -7 2 > 76 col. V 78
io6ob3i-io6iaio no 80 39 n.74 col. X 78
0.28 82 39 0.72 col. XI 78
io86b 3 0.2, 107 0.21 84 39 1111.72, 76, 40 col. XIV 770.86,
nn.8i, 83, and 89 78
Nicomachean Ethics
86 39 011.73-74, 76, col. XV 770.86
io96b 3 0.2, 107 0.21,
40 0.82 col. XVII 77
no 0.28
87 39 00.72-74, 76- col. XVIII 77
Poetics 77, 40 00.82, 85 col. XIX 770.87
i45&b 48 88 39 0.72 col. XXI 77, 78, 79
89 39 00.72-74 aod col. XXII 77, 78, 79
Rhetoric
76-77 col. XXIII 77
1415^5-17 46 0.14
90 39 011.72-74 col. XXVI 77
Sophistici Elenchi 93 39 00.73-76, 40
i73b 48 0.90 Diogeoes Laertius
94 39 0.72, 40 0.88 6.17 65 0.48
Topics
II mb 47 0.15 95 39 11.72, 40 00.81, 7-H7 53
89
96 39 00.72-74 aod Empedocles
[Aristotle]
76 DK 6 54 0.32
De Mundo 97 39 00.72-74 aod
401313-15 53 76 Euripides
98 39 0.72, 40 00.81, 8t Alcestis
Defixiooes 100 39 0.80 192-94 32 0.47
104 40 n. 81 224 29, 55
DTA (Wiiosch) 107 39 00.72 aod 76- 360 61
49 39 nn -7 2 > 76 77 634-50 63, 121
50 39 00.72, 76
51 39 00.76-78 SGD (Jordao) 658-72 121

52 39 00.72-73, 78, 40 i 39 00.72, 76 Andromache

0.82 3 40 0.90 17-20 103
53 39 00.72, 78, 40 11 40 0.90 105-6 22, 130 O.1O

n. 81 15 40 0.81
Bacchae
54 39 00.72-73 23 41 0.95
274-76 30, 53, 58
56 39 00.72, 76, 80, 40 40 00.81, 89
315 27 11.35
40 00.81, 89 44 40 0.90 508 17, 130 o.io
57 41 46 39 00.72, 76, 40 941-44 32 0.48
59 39 n-77 0.81 1197 18
60 39 00.73-74 51 40 0.90 1244 130 n.io
61 39 00.72, 79, 40 57 400.90
0.87 69 40 0.90 Cyclops
63 39 0.80, 40 0.81 95 39 0.72 649 27 0.35
64 40 11.81 99 39 0.72 Electra
65 39 0.77 100 39 0.72 812 32 0.48
66 39 00.72, 76 107 39 0.76 1275 103
68 39 00.72-74, 40 108 39 0.72
Hecuba
110.82-83 and 9° 124 40 0.83
342 32 0.46
69 40 00.81-82 146 41 0.95
70 40 0.82 Helen
71 40 00.82-83 Democritos 8-14 18-19, 71 n-67,
73 40 110.85-86 D K 2 69 101

74 39 nn-75-76> 4° DK 26 69 32-35 29, 54

00.82-83 and 85 DK 1223 69 149-50 104
75 39 0.72, 40 011.81-84 DK 142 69 241-51 29, 54
206 INDEX LOCORUM

1O
Euripides Iphigenia in Aulis 1493 21, 71 n.67, 9
321 18, 56 n.27
Helen (continued)
317 19 n. 18 332-63 57 1495 00 21

325-26 19 n. 18 558 270.35

679 32 n.47 Rhesus
530 19 n. 18 215-18 18, 101
818-23 19 n. 18 866 32 11.47
909-10 32 0.46 364 33
835-38 32 n. 46 894 18
1003 27 0.35 93° 27^35
1301-2 30, 53, 58, 71 1375-76 18 n. 15 Suppliants
11.67 1402 18 n. 15 885-87 18
1320 30, 53, 58, 71 11.67 1410-11 18 0.15, 27 888-89 17 n. 10
n
1327-28 30, 53, 58, 71 -35 1224-25 25
n. 67 1422-23 18 n. 15
1340 30, 53, 58, 71 n.67 1472 32 0.48 Trojan Women
2
i356 3°, 53- 58. ?i ".67 1595 18 n. 15 59° 7> 52
766-71 122
1581 32 0.48 Iphigenia in Tauris 989-90 l8, 31 11-44,
1670-75 25
32-33 !9> 71 n-67> 101 55, 63, 71 11.67, 101
Heracles 500 63 1168 27, 52
31 102 701-2 32 n. 46 1217 27, 52
808 61 944 36 n-528 n
1068-69 3 -46 Gorgias
Hippolytus 1454 104, 113
31-33 104
Encomium of Helen
79 27 n -35 Medea
2 22 n. 2O
307-10 18 n. 16 20-22 32
582-83 18 n. 16 495-97 V-
1131-34 18 n. 16 899 32 Heraclitus
1218-20 18 1070 32 D K 5 71-72
!343~45 27 0.35 DK 15 72
Ion DK 20 73
8-9 102, 103, 105, Orestes DK 33 74
114 37-38 35 n.57 DK2 5 73
29-30 1O2, 103, 1O5, 119-21 20, 55, 130 n. 9 DK 28 74
114 126 27 n. 35 DK 3 2 75
49-51 122 408-10 36 DK 36 73 11.72
69-73 ^2 n -53 557-59 120 n.48 DK 48 74
74-75 103 585-86 120 0.48 DK62 73,74
log-ll 122 0.53 954-56 20, 55, 130 n.g DK 88 73, 74
136-40 122 11.53 1008 104 DK 111 74 n. 76
240 27 n. 35
i635 3°' 53 DK 114 75
643 27 0.35 1646-47 103 DK 117 73 n. 72
661-63 71 n -67, DK 118 73 11.72
101-2 Phoenician Women
800-802 71 0.67, 25-27 16 DK 119 73
102 41-44 16
830-31 71 0.67, 102 150-52 15 n. 5 Herodotus
996-97 22 n. 21 395 27 11.35 1.7.3 1O3 n-1Q
1324-25 122 636-37 21, 24, 63, 7i 1.14.3 l°3> "3
1363 122 11.67 1.51.1 32 n.4$
1555-56 1O2, 103, 105 683-86 30, 53, 58 1.60.3 33 n-5 1
n. 16, 114 769 102 1.94.5-7 103 n.io
1575-78 103 801-5 *6 1.139 17 n-10
1581-88 101, 103 1104-9 J5 n-5 1.145 1O4
1590-94 103 1189-91 33 0.51 1.171.5-6 103 n.io
1.173.3 1O3 n-10
 INDEX LOCORUM 2O/

l.lSS.l 1O2 8.44.2 101, 103 5-77S-76 29> 54

2.30.1-2 34 n.53 8.124.2 33 0.51 5.844-45 29, 53
2.42.2 37 n.6o 8.136.1 102, 113 6.402-3 26-27, 51' 52
2.42.5 103 9.26.6-28.6 32 n.49, 35 7.138-41 25
2.46.2 36 n.6o 9.46.3 35 7.184 33
2.47.2 36 n.6o 9-47 35 9.116-17 28 n. 36
2.52.1-2 30, 54, 71 9.51.1-2 104 n.i3 9-236~37 33 n-51
0.67 9.90-92 28, 71 n.6j 10.274 33
2.61.1 36 10.542 32 n. 47
2.86.2 36, 37 n.6o Hesiod 11.450 18
2.98.2 103 12.200-209 34
2.132.2 36, 37 n.6o T/zeogony 12.215-29 34
2.144.2 37 n.6o i 67 12.237-40 34 11.52
2.170.1-171.1 36, 37 n.6o 116 ff. 51 n.25 13.821-23 33 11.51
3.55.2 102, 104, 114 144-45 !?> 71 n - 6 7 14.282 29, 54
4.6.2 102 182-85 35 0.56 14.290-91 50
~
4.45.2-4 103, 105 195-98 14, 31 n.44, 55, 2
71 n.6j, 104
°-73-74 50
4.52.1 104 11.13, 105, 20.215-17 103
113 0.33 199 15, 31 11.44 20.230 102, 114
207-10 21, 71 n.6j 20.231-32 103
4-59 63 O O £
4.107 26, 71 0.67 280-83 : 5> 71 n-°7 22.75 72
4.148.2-4 103 n.n 409 34 n.54 22.506-7 26, 51, 52
4.149.1 103 472 35 n.56
24.284-85 32
4.155.2-3 28 510-11 19, 101
24.292-95 33
4.161.2 28 535-60 19, 101 24.310-13 33
4.184.3-4 104, 113 11.33 565-67 19 24.319-20 33
4.198.1 104 11.14 775-76 58 n.35
O O£ O £i £ 24.499-501 26, 51, 52
4.204 105, 106 886-98 20, 56, 69
24.527-28 59
924 2O, 56
5.65.4 102 24.672 32 0.47
5.66.2 103 969-71 6l
24.727-30 26, 52
5.68.2 103 Works and Days
5.69.1 102, 113 1-4 29-30, 52, 67 Odyssey
5.926 16, 71 0.67 54 19, 101 1.1-10 67
6.47.1 103 n.n 80-82 21, 71 n.67 1.14-15 26 0.30
6.63.3 l^> 71 n -67 83-89 19 1.62 37 n. 61
6.103.4 102 1.70 28 0.36
213-73 15 n -3. 56 n-34>
6.111.1 32, 35 60 1.121 32 11.47
6.131.1 102, 113 256 15, 56 1.180-81 26
6.131.2 102 802-4 35 11.56 2.17 28 11.36
7.11.4 103 2.93-110 25
2
7.40.3 104, 113 Homer -i52~54 33
7.58.3 104 0.13 3.414 28 11.36
7.61.3 102 Iliad 4.499-503 17
7.62.1 103 11.10 1.1 67 4.509 17 11.13
7.74.1 103 n.io 1.403-4 50 5.99-102 26 n. 30
7.75.2 104 !-597 33 1 5.228-32 26
7.90-91 103 n.io 2.212-64 7 5.339-40 37 n.6i
7.92 103 n.io 2.262 72 5.423 37 n. 61
7.94 101, 103 2.339-41 32 11.46 8.43—45 19, 101
7.121.1 104, 113 2
-353 33 n-5 1 8.308-11 29 11.38
7.150.2 102 2.484 67 8.546 28 n. 36
7.178.2 103 n.n 2.811-14 5° 10.512-14 58 n. 35
7.193.2 25, 71 n.67 4-!55-59 32 n-46 11.421-30 20
7.231 18 n.i5 4474-77 104» H4 11.508-37 24
208 INDEX LOCORUM

Homer DK 8.36~3g 100 n.2 11.5-6 104

DK 8.53-54 10° n - 2 11.58 3411.54
Odyssey (continued)
DK 9 100 n.2
12.85-87 16
DKi7 31 n.45 Plato
13.330-32 19, 55, 102
DK 19 100 n.2
14.145-47 37, 38 Apology
15.160-65 33 223-C 146 11.56
Pindar
15.525-26 33
Charmides
15-531-34 33 n -5i Isthmian
i63b-d 47 n.i5
17.365-66 33 11.50 3 103 1750 100 n.2
18.5-7 102, 105 3.4-5 30, 53
18.258 3211.47 5.59-61 33 11.51 Cratylus
19.137-56 25 6-49-53 15. ?i 383-390 6, 81
19.260 36 n.67 383b 47
19.275-76 37 n.6i 7-8 103 3843 47 11.16
19.407-9 37 n.6i, 71 7.24 1O2, 112, 113 384b 65
11.67 3840-0 46, 47, 48
19.562-67 25 Nemean
3840-3856 8g
19.597 36 3.68 24 n.27
385e4-386dz 48
20.242-46 34 4.19 3411.54
38566 48
20.288-90 17, 71 11.67. 7-8 103
38653 145 n.54
101 7.6-8 24 n.27
3873 86
21.141-42 33 11.50 7.48 34 n.54
3886 84
23.19 36 7.61 24 n.27
3886-3906 85, 86
24.128-48 25 7.64-65 24 n.27
389-390 48 n.ig, 87,
24.199 20 7.85 34 n.54
144 0.51
24.311-13 33 11.51 7.86-89 24 n.27
38ga5-b4 88
8.47 3411.54
38gd 86
Homeric Hymns 11.20 34 n.54
3goe2-ii 84, 86
Hymn to Demeter Olympian 3god5~7 63 n.45, $6
18 61 2.7 34 n.54 39066-39109 48
489 61 5.19 103 39^-3973 85
6.30 16 n.8 3gib4~5 49 n.2o
Isocrates 6.45-47 16, 31 n.44 3gibg-c5 48
6-53-57 16, 31 n-44> 3gic 48, 49, 50, 52
Antidosis 7111.67 n.26
266-68 8 n.g 3gic-d 48, 4g, i6g
270-71 8 n.g 7 i°3
7.35-37 20, 56 3gid-3g2b 50, 87 n.ig
285 8 n.9 7.73-76 103 3gid2 5211.26
Busiris 9.63-64 102, 113 3916-3923 51
38-40 8 n.g 9.110-11 33 n.5i 39105-6 52 n.26
10.49-51 103 39235 52 n.26
Panathenaicus
17-34 8 n.g Pythian 392b-393b 50-5i
1.30-32 104 3g2bg 52 n.26
To Demonicus 2.90-92 72 392010 52 n.26
51-52 8 n.g 3g2d5 52 n.26
3-34 35 n -57
To Nicocles 39262 63 n.45
4 i°3
2-3 8 n.g 4.216 15 n.6 39262-39335 147
13 8 n.g 5.28 19 39254 52 n.i6
43-44 8 n.g 8 103 393-394 98> 123
48-50 8 n.g 8.42 25 n.28 39332 52 11.26
9 103 393a6-7 5i
Parmenides 9.5-8 103 393bi 51
DK 1.22-23 31 n-45 9.55-58 103 393b3~4 51- 52 n-26, 87
DK 8.17 100 n.2 9.68-69 103 3936 86
 INDEX LOCORUM 209

394b-c 51, 78 4O2b6 52 11.26 4i2b 6g, g4

3940 51, 63 n.45, 8o 4O2d8 61 4120 g4, 112 n.32
n.i, 87 n.ig 4O2dii~4O333 88 n.22 4i2c-4i3d 56, 60, g/)
394e-395a 84 4033 53, 61, 112 n.32, 41202 63 11.45
3g5a-b 51, 57 130 n.g 413 68
3g532~3 80 n.i 40335—4O4b4 88 n.22 41334-5 63 n.45, 68
3g5b-c 51, 88 n.22 4O4b 53, 61, 112 n.32, 11.63
3g5bz-8 80 n.i 130 n.g 4130-4143 g4
3g5b3 63 11.45 4040— d 60 n«3g, 130 41302 113 n^2
3g5b6 112 n^2 n.g 4143 6g
3g5c-d 51 40402-5 54 4i4b-c g4
3g5ci-2 56, 63 n.45 40405-7 62, 130 n.g 4i4b-e 88 n.23
3g5d-e 52 404d4 63 n.45 41405-6 52 11.26
395d3~5 63 n.45,8° n -i 4O4d8-e2 130 n.g 41532 52 11.26
395e-396b 5 2 > 53 4O4d8—40633 88 n.22 4i53g-e2 g5
3g565 87 n.ig 40401-2 55, 62 4i5b5 113 n.32
39631 63 11.45 40408-40534 55 4150-0 g4
3g632-bi 80 n.i 4o5b7 55 4i5d3 113 11.32
3g6sg 63 n.45 4°5b9 55 4i5d4 63 n.45
3g6bi 60 40501 63 n.45 41502 63 11.45
3g6b6~7 54, 60, 79 4O5d-4o6s 62, 130 n.g 4163 87 n.ig
3g6b8 63 11.45 40502-40633 55, 62 4i63io-dn 95
3g6c 52 n.26, 63 11.45, 40636 112 n^2 4i6b-~d 95
86 406)31-6 88 n.22 4i6bn 113 n.32
3g6d 66 4o6c-d 55 4160 87 n.ig
397b7 63 11.45 40606 63 0.45 4i6d8 63 n.45, U3
397c-d 54 40607 52 11.26, 5g 11.32
397c6 63 n.45 40601 63 n.45 41708 52 n.26
397d4-5 112 n.32 4073-0 56 4i7cg 113 n^2
3970-3980 68 40735 63 n.45 4i83-4igb 88 11.23
39702 63 11.45 4O73g 52 n.26 4i8d4 52 11.26
39705 52 11.26 4O7di— 3 56, 60 41802 63 n^5
3980-0 68 4O7d8—g 52 n.26 4igb7—01 g5
39801 112 n.32 40834 52 11.26 4igd7 63 n.45
39803 63 n.45 4o8bi 63 11.45 4igd8-e2 g5
3g8d6~7 86 n.i7 4o8c-d 52 11.26 4^0-4203 g3 11.32
399C5 63 n-45 4o8d-4ioe 131 11.14 42O3-b 88 n.23, 93
3ggd—4oob g4 4oga-c 68 n.32, g5
3ggdi-4oocio g5 4ogci 63 n.45 42ob7— g g5 n.42
39gd2-4oob7 88 n.22 40906 63 11.45 42od-e 6g
4003— b 68 40907 112 11.32 42137^1 g5
4oobi 63 n.45 4ogd—4iob 87 n.ig 42ib g5
4oob2 112 n.32 4iob7 63 n.45 42ib-c g5
4oob6 63 11.45 41002 52 n.26 42ic-d 87 n.ig
4oob8-cio 88 n.22 4iocg 63 n.45 42534-5 88 n.20
40004-5 63 11.45, 85 4100 86 425d~426b 88 n.2o
40134-6 87 4113 70 11.65, I3l n -H 425d5 52 11.26
4oibi-ei 88 11.22 4113-4210 65, 131 n.i4 426-427 gi
4oib8-g 85 4ns8-bi 131 11.14 4263 50 11.22, 87
40107 63 n.45 4iib-c 87, 88 4270 go
4oid3 112 n.32 4iib6 85 42831 52 11.26, 67
4oid7 63 n.45 4iid 6g, 94 42804-5 52 11.26, 67
4010 86 4iid-e 94 42807-8 67
40236 52 11.26 4110-4123 94 428d7-8 52 11.26
4O2b4 52 n.26 4123 88 11.22, 91, g4 42801-3 88
210 INDEX LOCORUM

Plato 5383 83 n.y looe-ioib 107

54105-6 83 n.y 10103-4 107
Cratylus (continued) io2b2-3 99, 114
5423 83 n.y
4310-4330 66 103-106 96 n.43
434-440 93 0.32 Laches io3b5 109, 115
4340-4350 89, 93 i85c-d 83 n.io io3b6-ci 99, 114
435aio-b3 94 igyb-d 47 n. 15 10303-7 111
435b-c 90, 94 1970 100 n.2 io3e9~io4b4 109 n. 26
435b2 90 1070-1153 92
Laws
435b5-6 94 6z6d3-5 111 n. 30 1093-1140 109
435C2-di go 10905 109
65436-7 8
43503 91
6800 149 n. 67 11033 1O9
43507 91 iiod3~4 109
81066-81137 8
4373 88 n. 22, 91, 94 11132 109
82ibg 113 n.32
438-440 56, 89, 90 11104 1O9
822b5 161
n. 28, 94, 96, 116, 170 me6 109
822by 161
4380 87 n. 19 ii3di 109
g333-c 38 11.65
439d 97, 98 11408 94 n. 34
g58eg 161
Critias 96305^7 113 n.32, 118 Phaedrus
n8b3 161 n. 41 2380 93 0.32, 113 0.32
Meno 244b-d 93 0.32
Euthydemus 24408 113 0.32
275c-d 67 75e 47 n-1?
ggc-d 146 n. 56 2453 146 n. 56
2770-2783 47 0.15 2510 93 11.32
28901-4 82 0.4 Parmenides 2550 93 n. 3 2
I28e-i2g3 3 n.2, 107 26704-6 50
Euthyphro n. 21 275b-C 171
6d-e 100 n. 2 i2gd 3 n.2, 107 n. 21
146 83 n. 10 278b-d 148 n. 63
i3ob-d 3 n.2, 107 11.21
i3oc-d 108 n. 22 Politicus
Corgias 2&3di 113 n.32
44809 82 11.4 13005-6 114 n. 34
i33di-2 114 n. 34 29409 85 n. 14
454 83 n.g, 133, 146 305b5 85 n. 14
n.55 i33d2~3 112 n.32
30502 85 n. 14
462b-c 82 n. 4 Phaedo 3ogdi 85 n. 14
462dg— 11 82 n. 4 6603 94 n. 34
463^-6 82 n.4 6832 94 n. 34 Protagoras
46402-46537 87 6837 94 n. 34 3^3-4 82 n.4
50161-7 82 0.4 69310 94 n. 34 31736 161
502 82 0.4, 130 n. 8, 6 b 31701-2 82 11.4
9 3-4 94 n-34 32338-9 82 n.4
132 72-77 4> !4° n-4°
5O3d-5O7c 142 0.45 74d-e 107 n. 21, 108 324dy~326b6 8 11.9
75c-d 144 32632 161
Hippias Major 3373-0 47
287d ff. 117 0.38 76012 94 n. 34
78d 3 11.2, 106, 145 33866-33931 48 n. 17
Ion 78d-e 107 n.2i 33931-3 48 n.iy
53ob 149 n. 67 7852 112, 113 34oa-b 47
34i3-b 47
533d-535a 146 0.56 ygc-d 107 n. 21
343b3 161
534b 134 79d7 94 n-34 358a-b 4 7 n.i 5
534b-d 83 n.y 8obi 96 11.43
53403 161 8od6-y 61 n. 41 Republic
535b-d 83 n. 7 ggb 68 11.63 33ob2-3 113
5356 83 n.y, 147 loob 3 n.2, 106, 145 33od-336a 137 11.32
535e~536d 146 0.56 looc 107, 115, 118 11.41 333-334 137 n -3 2
537d-e 83 n.y looe 115 342b4-en 132 n. 16
 INDEX LOCORUM 211

345b9~02 132 n.i6 44ib 150 0.70 5O9<i-5iie 83 n. 9,

346d5~ 34733 132 44id-e 154 n.8o 141
358d2 160 443cg-di 154 n.8o 5103 156
3(^5 160 443d4—07 154 n.8o 5ioby 140 n. 41
3630-3640 132 444-445 L54 5iib-c 140
36431 161 444d-e 159 n.8g 5iib6 140 11.41
364b-c 38 11.65 457c-d 124 0.58 5143 156 n. 83
365b-c 133 457d2~3 123 5143-5213 138
?,(>jdj 160 4590-4603 164 5i5b-c 130 n. 12
369C 93 45906 161 5i5c-d 157
3703-0 148 0.65 46ob7~ 8 123 5i6b 3 n. 2, 107, 145
372b 161 11.93, ^4 46ocg-di 123 5i6d 150 0.70
37602-9 8 46id-e ng, 124 5i7b-c 138, 144 11.49
3773-0 60, 150, 151 46234 85 n.i4 5i8c-d 138, 144 0.49
377d-e 151 4630-0 124 5213 1.38
377e-378b 58 11.36, 60 463cg-d6 123-24 52232-bi 163
378c-d 166 4&3d7 161 522b-54ib 163
378d-e 60, 150 4643 124 0.58 523-524 108
3793 148 465d— e 164 52502-3 no, 116
3790 58 468c~46ga 150 0.70 525d5~7 no, 116
3790-0 59 468d 161, 164 527cl-e 163
38008-9 58 474-480 10 11.17, 155 53005 85 n.i4
3833 149 0.67, 150 475d 9 n - n 5320 163
11.70 475d-48o3 83 n.g 533b 139, 143
3833-0 62 47509 95 n.4o 533b-d 140
3830 166 476-480 4, 130 n.i2 5330 133 n.ig
3863 166 47639^2 125 533e-534a Hi
3863-3870 57 47609^3 125 534b 133 11.19, 134, 163
387b 166 478d6—7 107 534d-e 86
387b-c 58, 59, 135 479a-b 125 538d8 85 n. 14
387b2~5 132 479d J-55 5403 163
3870 130 n.g, 163 4803 125 54ob-c 164
3890 149 0.67, 150 484b-e 133 56064 160
11.70 485a-b 133 56401 85 0.14
38ge-3go3 132 4gos-b 109-10, 116, 568b 160 nn. 91-92
3god 150 0.70 133 58od~58ic 155
3g2ci-4 158 n.88 493 !32 n-1?- 133 58od8 95 0.41
395d 15° 4930-4943 133 581010 160 n. 92, 161
3960-0 152, 165 495d i3g 5&3d8 160 0.92, 161
3g6di 162 4g6a 133 n.i8 58gb8 160
397d !32> !53> J59. l6° 4g8b 163 5953-b 165
11.91 5oob-d 138, 151 595a5 *53
3g8a-b 132, 157 5ood 138 595b-c 149 0.67
4oic-d i5g n.8g 5oib no, 115-16, 150 595b9 158 n.87
40104 149 n.6j n.7o 596 88 n.2i
4010-4023 61, 63, 151 5oib-e 138 596-598 144 n.5i, 155
4O2d i5g n.8g 5020-5033 163 5963 108
4030 i5g n.8g 5040 138 596b 144
42734 85 11.14 5O4d— 5ogd 138, 144 5g6bi2-di 137 n. 33
42gb~43ob 134 11.23, n.4g 596d-e 157
149 n.68 5053~5o6b 138 597b6 109 0.26, 115
433b8-ci g4 0.34 5o5d-e 138 597e !33
436b8-ci 165 n.gg 5080 i5g n.8g 59831 109 n.26, 115
43gd6-8 g5 n.4i 5080-5093 139 59837-10 145
439e-4403 g5 5ogb 138 598b 157
212 INDEX LOCORUM

Plato Symposium 597 120

ij-jbi 161 1145-56 120
Republic (continued)
i8od-e 79 n. 92 1194 120
5g8d 130 n. 8, 149 n. 67
18764-5 82 n.4
598d-e 133 Oedipus at Colonus
ig4d6 161
598d-6o2b 133 39-43 35. 58 n-35
ig6b-e 94 n. 36
59gb 160 nn. 91-92 58-65 103
5ggd 156 0.83 20232-9 94 n-34 107-8 102, 103, 105,
2106-2123 108
59962 85 0.14 114
6ooc 156 0.83 21064-5 1O9> U5 127-29 35
2113 97 0.45
6ooe-6oib 133 486-87 35, 58 n.35, 71
2iib 3 n.2, 106, 115, 145
6oia-b 133 n. 20 n.67
2i2b-e 161
6oib 130 n. 8 1130 32
223d5 82 n.4
6010-6023 144 1192 28
6o2b 159 n. 89 Theaetetus 1321-22 15
6020 164 17631 161 1323-26 28, 122
6020-6033 165 i8oe 100 n.2 1367-69 28, 122
6o3c-d 154 n. 79 1890-1903 92 1500-1503 23 n. 23
6o3d 165 n. 99
Timaeus Oedipus Tyrannus
6o3e-6o4d 165
406-413 51 n. 25 209-11 103
6o4b 165 0.99
47b4 161 674 27 n.35
6o4d-6o5c 165
5ib8 96 n.43 718-19 16
605-607 158
5234-5 112 11.32 919-21 29, 55
6050 130 n. 8
1031-36 16, 71 n.67
6o5c-6o6b 165
Plotinus 1349-50 16
6o6b5~7 165 n. 101
6o6c-d 165 Enneads Philoctetes
6o6e 156 0.83 2.9.16 148 n. 64 70-73 24
6066-6073 8 0.7, 149 5.8.1 148 11.64 79 27 11.35
0.67 245-47 24
6073 132, 152-53, 158- Prodicus 348-51 24
162, 164 nn. 97-98, DK 5 78 11.90 874 27 ^35
165 902 27 n. 35
6o7b 129 n. 6, 153 Sophoeles 93!-33 74
6o7c-d 149 n.6j 1229 21
60764-5 158 Ajax 1310 27 11.35
6o7d 132 430-32 17, 71 n.67 1344-47 18, 101
60701-2 143 516 35 n. 57 1413-33 18, 101
6o8b 143 549 27 n -35 Women ofTrachis
1005 17 n. 11
Sophist 807-12 120
1389-92 15 n.3, 56
22563 113 0.32 11.34, 60
817-18 120
225d4 113 11.32 926 35
22gd6 113 0.32 Electra 1064-67 121
25id ff. 118 n.4i 274 120 1181 32 n.4&
26702 113 n. 32 282-85 1O4 1199-1205 122